Fast Frameworks

The Alpha Engine Framework for Venture & Angel Investing

Don’t Hire an Agency. Build a Memetic Engine - The memetic engine framework

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

The Identity Migration Framework

The Energy Framework

Aditya Sehgal — Tue, 07 Apr 2026 18:54:17 GMT

A fully charged phone battery looks generous at 7am. By 2pm, with the screen on all day, three apps running in the background, and the brightness cranked up, it is at 11% and you are hunting for a cable. Nothing broke. You just spent faster than you recharged.

Your brain works the same way. You wake up with a full charge. By midday, if you have answered forty emails, sat through two meetings that could have been messages, made six decisions you will not remember by Friday, and let three notifications hijack your train of thought, your battery is nearly flat. And the actual work, the thing that matters, the call that requires real judgment, is scheduled for 3pm.

Anyone can work hard. Most people do. The ones who are still sharp at 3pm are not working harder. They are spending their energy on fewer, better things.

This framework is about how to spend your energy. You have a daily budget of peak hours and good decisions, and it is smaller than you think. You have four batteries, not one, and they drain and recharge independently. There are drains that do most of the damage, a kind of recovery that actually works (it is not rest), and one question about direction that decides whether all your other effort counts or gets wasted. The system runs on a simple rhythm: daily, weekly, monthly, quarterly. If you follow it, you will protect more of your capacity than most people realise they are wasting.

The Daily Budget

Most people never plan around it. You have a small handful of real peak hours per day, usually closer to two or three than to eight. That is the window where your brain can do its best thinking: solve hard problems, make judgment calls, produce original work, hold complexity in your head without dropping pieces.

If you are not sure which hours these are, use a simple test: they are the hours when you can read something dense and it stays in your head, when you can write cleanly without rereading every sentence, when you can hold five moving parts in your mind without reaching for notes. For most people, that is earlier than they want it to be.

Jim Loehr and Tony Schwartz, who spent decades studying elite performers from athletes to surgeons, found that the best performers do not work longer. They work in focused bursts of roughly 90 minutes, then recover. Two bursts is a good day. Three is exceptional. After that, cognitive quality drops off a cliff. You are still at your desk. You are still typing. But the email you draft in hour four rambles. The document you review misses things you would have caught at 9am. The decision you make at 4pm is the one you reverse on Tuesday. The output looks like work. The quality says otherwise.

Sustainable output = peak hours protected × decision quality maintained × recovery built in.

The same budget applies to your attention. Your brain can only process so much at once, and every task that needs focus competes for that space. Sophie Leroy’s research on attention residue showed that when you switch from one task to another, part of your mind stays stuck on the previous task for minutes afterward. You know the feeling: you leave a meeting, sit down to write, and it takes ten minutes before the writing makes sense again. On top of that, working memory holds about four things at once, and sustained attention gets worse by the hour whether you feel it or not. You only have so much, and it breaks apart when you load too much onto it.

There is also evidence that willpower itself works like a fuel tank. In practice, every decision you make, from what to eat for breakfast to whether to approve a hire, seems to draw from the same pool. The original studies overstated the size of the effect, but the practical observation holds: your early decisions tend to be clean and your late ones tend to be sloppy. Whether that is because your willpower ran dry or because your attention got overloaded, the result is the same. You have a finite number of good decisions per day, and the number is smaller than you think.

In my Decision Framework, I describe the Impact Gate: a filter that sorts decisions by whether they actually matter before you spend any energy on them. Most decisions do not get through the gate. That is the point. Every low-stakes decision you let through uses up attention that your high-stakes decisions need. AI has made this worse, not better. More tools means more options, and more options means more decisions. I will come back to this in the Drain Map.

Put these together and you get a planning default for the daily energy budget. A few peak hours. A handful of real decisions. Everything else should either be on autopilot (habit), delegated, or eliminated. The person who fills their best hours with email and meetings and saves the hard thinking for 4pm is not disciplined. They are wasting their most expensive resource on their cheapest tasks. And none of this matters if the ladder is leaning against the wrong wall.

Four Batteries, Not One

For years I treated energy as a single tank. When I was tired, I rested. When rest did not work, I rested harder. I took weekends off, I slept more, I went on holiday. And I would come back and within three days feel exactly as depleted as before. The rest was real. The recovery was not. I was refuelling the wrong tank.

Loehr and Schwartz identified four types of energy. The important part is that they drain and recharge independently. You can be physically fine but emotionally wrecked. Mentally sharp but spiritually empty. Understanding which battery is low changes what kind of recovery you need. Reaching for the wrong one feels worse than doing nothing, because you did the “right thing” and nothing changed.

A quick diagnostic makes the difference between a true fix and a placebo.

Physical energy is the foundation. Sleep, movement, nutrition. If you are sleeping six hours when your body wants eight, no amount of willpower or coffee will give you back what the missing sleep took. The research is consistent: one week of sleeping six hours a night leaves your brain performing as badly as if you had stayed awake for 48 hours straight. You do not feel impaired. You are. Physical energy is the battery that powers all the others.

Emotional energy is the quality of your internal state. Frustration, resentment, anxiety, and conflict drain this battery fast. A difficult conversation with a colleague can cost you more energy than four hours of focused work. Emily and Amelia Nagoski showed that stress and the stressor are different things. You can remove the stressor (finish the project, end the argument) and still carry the stress because the body’s stress cycle was never completed. The cycle completes through physical discharge: exercise, laughter, tears, physical connection. If you skip the discharge, the stress stays in your body even after the cause is gone.

Mental energy is focus and concentration. This is the battery that holds the two to three peak hours. It drains when too many things compete for your attention: too many open tasks, too many decisions, too much jumping between things. Every time you switch from one task to another, your brain pays a switching cost. It needs time to load the new context and let go of the old one. Three switches in an hour can destroy the entire hour. AI tools often make this worse, not better. If an AI tool interrupts your focus to save you time, it costs you more than it saves.

Spiritual energy is purpose and direction. Not religion. The feeling that your effort is connected to something that matters to you. In my Identity Migration Framework, I describe how people whose identity sits at the purpose level can absorb setbacks that would crush someone anchored at the task level. Purpose energy is what keeps you going when the other three batteries are low. It is also the slowest to recharge when it drains, which is why burnout that reaches the purpose level takes months or years to recover from, not days. AI is pressing on this battery from two directions. For driven people, it creates anxiety: expectations rise faster than achievement, and you feel like you are falling behind no matter how hard you work. For disengaged people, it creates apathy: everything becomes easy to produce, nothing feels worth producing. Both drain the spiritual battery. Both feel like personal failure. Neither is.

That diagnostic changed everything for me. Once I could name which battery was low, the right recovery became obvious. When I was physically exhausted, I needed sleep, not a pep talk. When I was emotionally drained, I needed connection, not a day off. When I was mentally fried, I needed a walk in the woods, not another podcast. When I had lost my sense of purpose, I needed to revisit the wall my ladder was leaning against, not optimise my morning routine. Every time I had tried to rest my way out of exhaustion and failed, it was because I was recharging the wrong battery.

Flow vs Longevity

Most productivity writing treats flow as the goal. Get into flow. Stay in flow. Protect your flow state. There is a cost to this that the productivity writers leave out.

In my Life Quality Framework, I define life quality as attention divided by time. The richer your experience, the more attention you were paying. When you are fully present, time feels long and full. A week on holiday in a new country feels like a month because everything is novel and your brain is paying attention to all of it. A week at your desk in flow feels like an afternoon because your brain has compressed everything into a single focused stream.

Flow is wonderful for output. It is terrible for the felt length of your life.

This is not to say flow is bad. Flow feels meaningful in the moment. It can be deeply satisfying. Effort and skill align, time stops mattering, and the work feels like it matters. The problem is flow as a default mode of living. In that mode, you get a lot done and experience very little of it. The day disappears. You look up and it is dark outside and you cannot remember eating lunch. That is useful for the two to three hours of peak work. It is a disaster if it consumes your entire waking life. The person who lives in permanent flow gets a great deal done and, at the end of the year, feels like the year lasted three months.

The operating system here treats flow as a tool, not a lifestyle. Use flow for the peak hours. Then come out of it deliberately and pay attention to the rest of the day. Walk somewhere without headphones. Eat a meal without a screen. Have a conversation where you are not thinking about what you need to do next. That attention is what makes a life feel long and full. Flow compresses. Presence expands. You need both, on purpose, not by accident.

The Drain Map

Four drains do most of the damage.

Decision volume. Too many items on the runway. What to wear. Which meeting to accept. How to respond to an email that does not matter. Each one costs almost nothing on its own. Added up over a day, they cost you everything. Eliminate, automate, or delegate as many as you can. Wear the same thing. Batch your email to two slots. Set default rules for meeting acceptance. Each one you remove from the queue saves attention for the work that actually requires your judgment. In my Time Mastery Framework, I describe how the first two hours of your day should be reserved for your Big Thing, with no meetings, no email, no internet. That is not a productivity hack. It is battery protection.

AI saves energy only if you use it to eliminate decisions, not to create more of them. An AI that auto-files your low-priority email saves you a decision. An AI that shows you three draft replies and asks you to pick one adds a decision. The test is always: fewer decisions, or more?

Interruptions. Decision volume is too many items on the runway. Interruptions are forced mid-flight diversions. A quick glance at your phone. A Slack message. A colleague asking “got a minute?” Each one yanks you out of whatever you were doing and forces a reload. Your brain does not cleanly release the previous task. It lingers. Three interruptions in an hour can destroy the entire hour. Each one is tiny. But ten tiny interruptions in your best hour and the hour is gone. Your best work gets eaten by things that took ten seconds each.

Identity erosion. This is what spiritual battery drain feels like in modern life. If you notice that you are working harder but caring less, or caring more but producing less, the spiritual battery is the one running flat. In my Identity Migration Framework, I map how this pattern is accelerating. The anxiety version: you watch AI-generated output flood your field and feel like you are falling behind no matter how fast you run. The apathy version: AI makes everything easy to produce and nothing feels worth the effort anymore. Both are energy drains. Both end at the same place. Working harder does not fix either one.

Recovery Is Not Rest

Most people think recovery means sitting on the sofa. It does not.

Hans Selye, the endocrinologist who founded modern stress science, mapped the body’s response to sustained pressure. The pattern has three phases. First, alarm: the body activates, cortisol spikes, you feel the stress. Second, resistance: the body adapts, you push through, you feel like you have adjusted. Third, exhaustion: the body’s reserves run out and systems start failing. The dangerous part is that the resistance phase feels like strength. You think you are handling it. You are actually spending down a finite reserve, and the transition from resistance to exhaustion is a cliff, not a slope. You do not feel it coming. You just fall off.

Recovery has to be deliberate because the body does not automatically complete the stress cycle. The Nagoskis showed that stress creates a loop in the body that needs closing. The stressor might end (the deadline passes, the argument resolves) but the stress chemicals are still circulating. They stay until the body discharges them through physical action. Exercise is the most reliable discharge. Laughter, crying, physical affection, and creative expression also work. Scrolling your phone does not. Watching television does not. Sitting still and “relaxing” does not close the cycle. The stress stays in the body, accumulates, and eventually tips you from resistance into exhaustion.

Recovery is active. It requires doing something, not doing nothing.

Each of the four batteries recharges differently. Physical energy recharges through sleep, movement, and nutrition. Emotional energy recharges through connection, laughter, and completing stress cycles. Mental energy recharges through downtime, nature, and activities that absorb you without demanding focus (cooking, walking, music). Spiritual energy recharges through purpose, contribution, and anything that reconnects you to why you are doing what you are doing.

If you are depleted tonight and have twenty minutes, here is the lowest-effort version of active recovery for each battery. Physical: a twenty-minute walk outside, no destination. Emotional: call one friend and talk about something that is not work. Mental: cook something simple with no screen running. Spiritual: write one sentence about what you want this work to do for someone else. None of these require motivation. They require a decision to start, and the decision takes less energy than staying on the sofa pretending to rest.

The Wall Question

In my Ladder and Wall Framework, I describe how your career has two parts: the ladder (your skill, the thing you climb) and the wall (the direction, the thing the ladder leans against). Most energy advice is about climbing the ladder more efficiently. Sharpen your focus. Protect your peak hours. Batch your decisions. That is all useful. But the single energy decision that matters most is not about the ladder. It is about the wall.

If the wall is wrong, all the energy management in the world is wasted. You are climbing efficiently in a direction you did not choose and do not want. The spiritual battery drains not because you are working too hard but because the work is taking you somewhere that does not matter to you. That is the kind of fatigue that no amount of sleep, exercise, or time management can fix.

The Ladder and Wall Framework describes platforms: the flat spaces between climbs where you stop and look around. While you are climbing, your attention is narrow and upward. You cannot see the wall clearly. On a platform, you get distance. You can ask: is this wall still the right one? Is the energy I am spending taking me somewhere I actually want to arrive?

If the answer is no, the most energy-efficient thing you can do is change walls. It feels like a loss in the short term. Every wall change resets your position on the ladder. But a person climbing the right wall at half speed will arrive somewhere meaningful. A person sprinting up the wrong wall will arrive somewhere meaningless, faster. If the wall is right, every hour you protect and every drain you cut makes your life better. If the wall is wrong, none of it matters.

Burnout: The Cliff You Cannot See

Herbert Freudenberger coined the term “burnout” in the 1970s studying volunteer clinic workers. His original description had three signs: exhaustion (the body gives out), cynicism (you stop caring about the work), and reduced efficacy (you keep working but the results get worse). All three have to be present. Being tired is not burnout. Being cynical is not burnout. The combination of all three, sustained over weeks, is.

The reason burnout is dangerous is Selye’s cliff. In the resistance phase, you feel functional. You are tired but you are managing. The slide from resistance to exhaustion happens fast and without warning. One week you are grinding but coping. The next week you cannot get out of bed, or you find yourself staring at a screen for twenty minutes without processing a single word. The cliff is invisible from the resistance phase. You only see it after you have gone over it.

Three early warning signs, before the cliff:

Recovery stops working. A weekend used to recharge you. Now it does not. Sleep does not feel restorative. Exercise feels like another obligation. The activities that used to refill your batteries no longer do. This means your reserves are running low and the exhaustion phase is approaching.

Small things trigger big reactions. A minor request from a colleague makes you furious. A routine email feels overwhelming. The emotional battery is so low that things which would normally be nothing now feel like everything.

You cannot remember why you started. The purpose that used to fuel the work has gone quiet. You can still describe it if someone asks, but it does not generate energy anymore. The spiritual battery is flat. This is the deepest kind of empty, and the slowest to come back from.

If you recognise these signs, the right response is not to push through. It is to stop climbing and find a platform. Revisit the wall. Recharge the specific battery that is lowest. And give it more time than you think it needs, because the damage from the resistance phase builds up invisibly.

The Complete Picture

The budget: A few peak hours. A handful of real decisions. Everything else should be on autopilot, delegated, or eliminated.

The batteries: Physical (sleep, movement, nutrition), emotional (connection, stress cycle completion), mental (focus, protected from interruption), spiritual (purpose, the wall question). They drain and recharge independently. The fix depends on which one is low, and most people reach for the wrong one.

The trade-off: Flow is for output. Presence is for living. Use flow for the peak hours. Come out of it deliberately for everything else.

The drains: Decision volume, interruptions, identity erosion. Cut the first two. Monitor the third. Test every tool, including AI, by whether it leaves you with fewer decisions or more.

Recovery: Active, not passive. Match the recovery to the battery that is lowest. Stress cycles must be closed through physical discharge, not waited out.

The wall: The most important energy decision is which direction you are climbing in. If the wall is wrong, efficient climbing makes things worse, not better.

Burnout: Three signs: exhaustion, cynicism, reduced efficacy. The cliff between resistance and exhaustion is invisible. Watch for recovery failing, small triggers causing big reactions, and purpose going silent.

The Operating System

This is the smallest set of moves that keeps all four batteries from discharging.

Protect the budget. Recharge the batteries. Watch the wall. When something feels off, run this diagnostic: if your output is dropping, check the mental battery and the drains. If your mood is dropping, check the emotional battery and whether you are completing your stress cycles. If your motivation is dropping, check the spiritual battery and the wall.

Daily: protect the peak. Block your best hours for the hardest work. No meetings, no email, no notifications. Treat this block the way you would treat an appointment with someone important, because it is. Batch everything else into the remaining hours. In my Time Mastery Framework, I describe how 300 working days at two hours per day gives you 600 hours of focused work per year. Very few goals survive 600 focused hours. The problem is never the total time available. The problem is how much of it you protect.

Daily: close the stress cycle. Move your body every day. It does not have to be a gym session. A twenty-minute walk works. The point is to give the stress chemicals a physical exit. If you skip this, the stress accumulates regardless of whether the stressor is still present.

Weekly: check the batteries. Take five minutes at the end of each week and ask: which battery is lowest? Physical, emotional, mental, or spiritual? Plan the weekend to recharge that one specifically. A weekend that recharges the wrong battery feels restful but leaves you depleted in the area that actually matters.

Monthly: audit the drains. Look at the decisions, interruptions, and commitments that consumed your energy this month. Which ones were necessary? Which ones can be eliminated, automated, or delegated? Each unnecessary drain you remove frees up space for the things that matter. This compounds. Small reductions in waste, repeated monthly, add up to a big difference over a year.

Quarterly: check the wall. Step onto the platform. Look around. Is the wall still right? Is the energy you are spending taking you somewhere you want to arrive? If the answer is yes, keep climbing. If the answer is uncertain, stay on the platform until it becomes clear. If the answer is no, change walls. Changing walls costs a lot up front. It is the best investment you will ever make.

Applications

Early career: Your batteries are large and they recharge fast. The danger is not burnout but misdirection. You have so much energy that you can climb any wall at speed, which makes it easy to climb the wrong one for years before you notice. Use the quarterly wall check early. It is cheaper to change walls at 25 than at 45. In my Promotion Flywheel Framework, I describe how the Big Thing requires two protected hours per day. That is not just a career strategy. It is an energy strategy. The two hours protect your best output from being consumed by other people’s priorities.

Mid-career: Your batteries are smaller than they were and they recharge slower. The danger is the resistance phase: you feel like you are coping, but your reserves are quietly running out. Mid-career is where most burnout happens, because the demands have risen (leadership, family, financial obligations) while the energy supply has not. Audit the drains ruthlessly. Protect the peak hours. And take the wall question seriously. Many mid-career crises are not about workload. They are about climbing a wall that stopped mattering five years ago and running out of spiritual energy as a result.

Leadership: Your job has shifted from doing the work to making decisions about the work. That means your daily decision budget is the bottleneck. Every unnecessary decision that reaches your desk is a tax on the decisions that matter. Build systems that handle routine decisions without you: clear delegation, decision frameworks your team can apply, defaults that only escalate when they should. In my Team Design Framework (to be published), I describe how a well-designed team structure is also an energy structure: it routes decisions to the people with the most context, preserving your capacity for the decisions only you can make.

Career transitions: Changing walls drains all four batteries simultaneously. You are physically stressed (new routines, disrupted sleep), emotionally exposed (new relationships, loss of old ones), mentally overloaded (everything is unfamiliar), and spiritually uncertain (you left the old purpose and the new one has not yet formed). Budget for this. Transitions take more energy than normal climbing, and most people underestimate the cost by two or three times. Build extra recovery into the transition period. Do not expect your normal output levels for the first six months.

Parenting: Children drain the emotional and physical batteries relentlessly and recharge the spiritual battery in ways nothing else can. The danger is that parents sacrifice all recovery time, treating it as selfish. It is not selfish. A parent running on empty is less patient, less present, and less able to give the attention that children need. Protecting your own recovery is not a luxury. You cannot give what you do not have. And the flow-versus-presence trade-off matters here more than anywhere: the years with young children go fast precisely because parents spend them in survival flow. Slow down. Pay attention. Those years will feel longer and richer for it.

You woke up this morning with a full charge. Tomorrow you will wake up with a full charge again. You can recharge overnight. You cannot get the day back. Protect the peak hours. Close the stress cycle. Check the wall. And every now and then, put the phone in your pocket, take your headphones out, and walk somewhere slowly enough to remember it.

Bibliography:

Baumeister, R.F., Bratslavsky, E., Muraven, M., & Tice, D.M. (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology, 74(5), 1252–1265. doi.org/10.1037//0022-3514.74.5.1252

Cowan, N. (2001). The magical number 4 in short-term memory: A reconsideration of mental storage capacity. Behavioral and Brain Sciences, 24(1), 87–114. doi.org/10.1017/s0140525x01003922

Freudenberger, H.J. (1974). Staff burn-out. Journal of Social Issues, 30(1), 159–165. doi.org/10.1111/j.1540-4560.1974.tb00706.x

Leroy, S. (2009). Why is it so hard to do my work? The challenge of attention residue when switching between work tasks. Organizational Behavior and Human Decision Processes, 109(2), 168–181. doi.org/10.1016/j.obhdp.2009.04.002

Loehr, J., & Schwartz, T. (2003). The Power of Full Engagement: Managing Energy, Not Time, Is the Key to High Performance and Personal Renewal. Free Press. simonandschuster.com

Mackworth, N.H. (1948). The breakdown of vigilance during prolonged visual search. Quarterly Journal of Experimental Psychology, 1, 6–21. doi.org/10.1080/17470214808416738

Nagoski, E., & Nagoski, A. (2019). Burnout: The Secret to Unlocking the Stress Cycle. Ballantine Books. penguinrandomhouse.com

Selye, H. (1936). A syndrome produced by diverse nocuous agents. Nature, 138, 32. nature.com/articles/138032a0

Check out some of my other Frameworks on the Fast Frameworks Substack:

The Thirty-Three Kilometre War

The Alpha Engine Framework for Venture & Angel Investing

Don’t Hire an Agency. Build a Memetic Engine - The memetic engine framework

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

The Thirty-Three Kilometre War

The Identity Migration Framework: Moving your sense of self to higher ground in the age of AI

Aditya Sehgal — Thu, 26 Mar 2026 00:45:11 GMT

There are two kinds of people at dinner parties. The ones who answer “what do you do?” with tasks they perform, and the ones who answer with problems they solve. The first group is about to have a very difficult decade.

The comfortable middle-class knowledge work that built careers for the last thirty years is being automated faster than anyone expected. ChatGPT can write the quarterly report. Claude can build the financial model. Midjourney can design the presentation. The question is not whether machines will change your job. The question is whether your sense of self can survive that change.

The Career Ladder: Four Levels of Identity

When someone asks what you do, your answer points to one of four things. Each represents a different relationship with automation. Each requires a different strategy for survival.

Level 1: Task Identity

“I write the quarterly report.”

You identify with the artifact. The spreadsheet, the presentation, the code, the design. When the thing you make is good, you feel good. When someone else can make it faster, you feel threatened.

Machines live here. They produce artifacts. If your identity is anchored at this level, you and machines are competing for the same work. You will lose that competition, and it will feel personal.

Sarah spent eight years as a marketing coordinator. She took pride in her email campaigns. She loved testing subject lines, setting up customer groups, tracking open rates. Then her company bought software that could test a hundred variations in the time it took her to write three. The campaigns got better. Sarah felt invisible. Her identity was the campaigns. When software could make them better, what was left of her?

In my AI-Jobs Framework, I mapped roles across two dimensions: how high the stakes are, and how repetitive or creative the work is. The roles at highest risk are repetitive, low-stakes tasks. These are task-identity roles almost by definition. The work is the artifact. The person is the pair of hands.

Level 2: Responsibility Identity

“I make sure we enter the right market.”

You identify with the outcome, not the artifact. You might still write the report, but the report serves a decision. You own whether that decision was right. You sleep well when the market entry succeeds and poorly when it fails.

Machines can write the report. They cannot own the consequence. There is no computer to blame when the strategy fails, no algorithm that loses reputation, no machine that lies awake wondering what it missed.

This is the first level where automation becomes a tool rather than a competitor. You use it to produce better artifacts faster, then stake your professional life on what those artifacts recommend.

This is the difference between being the architect and the bystander. Responsibility identity is what makes you the architect.

Level 3: Capability Identity

“I spot the patterns that change the strategy.”

You identify with accumulated judgment. The instincts you have built over years of making calls and watching what happened. The patterns you learned to see that others miss. The decisions you got right when the data was unclear.

Marcus runs revenue operations for a software company. He can look at a month’s worth of sales data and tell you which deals will close, which salespeople are burning out, and which market segment is about to shift. He cannot explain exactly how he knows. Fifteen years of watching sales cycles taught him to read signals that do not show up in any dashboard.

In my Xeme Framework, I introduced the idea of a xeme: the smallest piece of learning earned through real-world testing and built into how you work. Your capability identity is really the collection of xemes you carry. The decisions you got right and wrong. The patterns you spotted early. The instincts nobody taught you because they can only come from experience.

Machines can copy the outputs of judgment. They cannot copy the judgment itself. Your pattern recognition is yours. Your instincts come from choices you made and consequences you lived through. Those are not transferable to a machine, no matter how much data you feed it.

Level 4: Purpose Identity

“I help organisations make better decisions, even when it’s hard.”

You identify with the problem you chose to solve. Tools change, industries change, roles change. But the problem persists, and you persist with it. You could switch careers tomorrow without an identity crisis because your identity was never attached to any particular career.

The key word is “chose.” Purpose is not a direction someone gave you. It is a commitment you maintain when you could walk away. You wake up on a difficult Monday, consider quitting, and choose to keep going because the problem still matters to you.

Machines can be given a purpose. But they cannot choose one. They were never going to stop. They have no alternative they rejected, no cost they are paying, no Monday morning decision to keep going. The identity value of purpose comes from the fact that you could walk away and you do not.

In my Ladder and Wall Framework, I describe how your career has two parts: the ladder (your role, the thing you get better at) and the wall (the industry, the direction the ladder leans against). Purpose identity is what lets you change walls without losing yourself. If your purpose is “I help organisations make better decisions,” you can move from consulting to tech to education and your identity travels with you. Task identity cannot survive a wall change. Purpose identity does not even notice it.

How to Read the Ladder

Each level up fundamentally changes your relationship with automation. At the Task level, you find yourself competing directly with machines for the same work. Move to Responsibility, and those same machines become your assistants, helping you deliver better outcomes you own. At Capability level, automation amplifies your reach while your hard-earned judgment remains distinctly yours. And at Purpose level, machines become irrelevant to your identity entirely. You pick them up or put them down like any other tool.

Most people move between levels depending on the situation. You might operate at Capability level in your main role while staying at Task level for a side project. What matters is where your primary sense of self lives. That anchor determines whether automation strengthens or weakens your happiness.

The Migration Path

The movement is not a single jump. It is a series of shifts in what you feel responsible for.

Task to Responsibility: Stop identifying with the artifact. Start identifying with the outcome. Ask: “If this thing I make disappeared tomorrow, what would actually be lost?” If the answer is “the decision would be worse,” then the decision is your real work.

Responsibility to Capability: Stop identifying with any single outcome. Start identifying with patterns across outcomes. Ask: “What do I know now that I did not know five years ago, and how did I learn it?” That knowledge is your real asset.

Capability to Purpose: Stop identifying with your thinking. Start identifying with the problem you care about solving. Ask: “If I had to start over in a completely different industry, what problem would I still want to work on?” That problem is your purpose.

Each move takes time. The further your current role is from judgment-based work, the longer the migration. But every step up is permanent. Once you own an outcome, you do not go back to owning an artifact. Once you see patterns, you do not unsee them.

The Life Anchors: Three Sources Machines Cannot Touch

Career identity is not the whole story. Part of your sense of self lives in places machines cannot reach, no matter how advanced they get. These are your life anchors. They protect your happiness even if the career ladder gets disrupted entirely.

Presence: Being There

Every experience that requires you to show up physically, emotionally, completely. Saturday morning pancakes with your daughter, where you are the only one who knows she likes the burnt edges. Sitting with a friend whose marriage is falling apart and not trying to fix it, just being there. Running the last mile of a race when your legs feel like concrete.

Computers can process information about these moments. They cannot be in them. The value comes from the fact that you were there and it would not have been the same without you.

Stakes: Having Something to Lose

Every experience where the outcome is uncertain and the cost of failure is personal. Starting a business with your savings. Performing on stage when you might forget your lines. Telling your partner you are scared about money when you have always been the stable one.

Machines do not risk anything. They have no skin in the game. The identity value comes from choosing to do something when you could lose and it would matter to you.

Meaning: Choosing What It All Means

Making sense of what happened and deciding what it means for you. Computers can analyse patterns in your behaviour, but you decide whether your career pivot was brave or reckless. Whether your relationship challenges made you stronger or more guarded. Whether your failures were education or evidence.

The interpretation is yours. The story you tell yourself about your life is yours. Machines might offer data about your patterns, but the meaning you extract from that data lives where your identity lives.

The Thread: Kindness

In my original Happiness Framework, I wrote that kindness is the simplest and most powerful happiness hack. Be kind. It gives you achievement with almost zero expectations, and that combination drives a positive loop that lifts your entire happiness set-point.

In this framework, kindness plays an even deeper role. It works at every level of the career ladder and through every life anchor.

Task-level kindness: helping a colleague with their presentation when you did not have to. Responsibility-level kindness: taking accountability for a team failure even when it was not entirely your fault. Capability-level kindness: sharing pattern recognition that helps others succeed. Purpose-level kindness: choosing to solve problems that make other people’s lives better.

Kindness through presence: showing up for someone when you could have stayed home. Kindness through stakes: giving something up for someone else’s benefit. Kindness through meaning: helping others make sense of their struggles.

Kindness is the one move that builds identity regardless of where you are on any dimension. If you do nothing else, be kind. It works everywhere.

The Complete Framework

Vertical axis (career): Task → Responsibility → Capability → Purpose. Each level up makes automation more useful and less threatening to your identity. Move up by shifting what you feel you own: from artifacts, to outcomes, to patterns, to problems.

Horizontal axis (life): Presence, Stakes and Meaning. Three sources of identity machines cannot touch. Build all three while you have time.

Thread: Kindness. The single action that builds identity across every level and all three anchors.

A Migration Story

Elena spent twelve years as a financial analyst. She built complicated models, ran scenario analyses, created presentations that executive teams used to make billion-dollar decisions. She was good at her work. She identified completely with being good at it.

Then her company started using software that could build models faster than she could think through the assumptions. The recommendations got better. The executives still needed someone to translate machine output into human strategy, but Elena felt like a middleman in her own career.

The migration took eighteen months. First, she stopped caring whether she personally built the model and started caring whether the model led to good decisions. When a recommendation based on her analysis helped avoid a costly acquisition, she owned that win differently than she had owned her previous modeling accuracy.

Next, she began noticing patterns the software missed. The model would flag Company A as a good acquisition target, but Elena had seen three deals like this before and knew the cultural integration would fail. Her judgment became her differentiator, not her technical skill.

Finally, she realised what she actually cared about: helping companies spend money on things that matter. The industry became secondary. The tools became secondary. The problem became primary.

Today Elena runs strategy for a climate tech firm. She uses software for modeling. She owns the outcomes. She trusts her judgment. Her identity survived the transition because she moved it to higher ground before the flood.

Applications

Early career: If you are three years in and still identifying with deliverables, you are building toward a career that automation will hollow out. Start the migration now while you have time. Ask for ownership of outcomes, not just tasks.

Mid-career: This is where the fork matters most. Fifteen years of deepening task skills puts your identity in automation’s crosshairs. Move to responsibility level immediately. Ask your boss for accountability, not more work.

Leadership: Your job is to help your team migrate. Give people ownership of outcomes, not just deliverables. The leader who says “write me a report by Friday” anchors their team at task level. The leader who says “figure out whether we should enter this market and come back with a recommendation” pushes them toward responsibility.

Parenting: Children build identity the same way adults do. If everything is done for them (by parents or by machines), they stay at task level or below. Let them own outcomes. Let them fail. Let them experience presence, stakes, and meaning early and often. These are the raw materials of resilient identity.

Retirement: When the career ladder disappears, only the life anchors remain. People who retire with strong presence, stakes, and meaning transition well. People whose entire identity lived on the career ladder struggle, regardless of which level they reached. Build the horizontal anchors before you step off the vertical ladder.

The framework is simple. The migration is not. But the alternative (watching automation erode an identity you spent decades building) is worse.

Move your sense of self to higher ground. The flood is coming. The high ground is beautiful.

Check out some of my other Frameworks on the Fast Frameworks Substack:

The Alpha Engine Framework for Venture & Angel Investing

Don’t Hire an Agency. Build a Memetic Engine - The memetic engine framework

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

The Thirty-Three Kilometre War

Aditya Sehgal — Fri, 20 Mar 2026 19:40:59 GMT

Editors note: This is fiction. The frameworks are real

The most consequential weapon in the Iran conflict was never nuclear. It was geographic.

The Strait of Hormuz is 33 kilometres wide at its narrowest point. Twenty percent of the world’s oil passes through it every day. Iran does not need to close it. It needs Lloyd’s of London to believe it might.

The opening looks the way everyone expects. Cruise missiles flatten enrichment sites at Natanz, Isfahan, Fordow. Satellite imagery circulates within minutes. Cable news declares mission accomplished before the second wave has launched. But Iran’s nuclear programme was never a building. It was a network, dispersed and buried and duplicated across sites that Western intelligence mapped and sites it did not. The strikes destroy roughly 60% of visible infrastructure. Sounds decisive until you understand the remaining 40% was the portion Tehran expected to lose.

The side that plans for the punch absorbs it better than the side that throws it.

By day five, the Pentagon briefs “significant degradation.” Tehran briefs the programme is intact. Both are partially true. Neither side has a reason to stop. And then the real war begins, on the water.

Just before dawn, a swarm of cheap drones crosses low over the Gulf. Too small to terrify on radar. Too many to ignore. They hit compressors, loading gantries, a desalination intake. Nothing world-ending. Precision is expensive. Nuisance, repeated often enough, does almost the same job. Insurance markets did the rest. Three tankers turned around within ninety minutes of each other. None of them had been hit. All three feared they were about to be. Oil moved to $140, freight rates spiked, and in one trading room in Singapore a junior analyst said it plainly: “This is not a supply problem. This is a courage problem.”

He was right. Modern systems do not always break when something is destroyed. Sometimes they break when enough people decide the crossing is no longer worth the risk.

The actor who controls the price of uncertainty has more leverage than the actor who controls territory.

Within the same fortnight, rockets hit US bases in Iraq. Houthi drones target Saudi oil infrastructure. Hezbollah fires a calibrated barrage into northern Israel, enough to force a response, not enough to trigger a ground invasion. Four theatres, none decisive on their own, all of them consuming diplomatic bandwidth, military assets, and political capital at once. In Dhaka, a minister began drafting food subsidies he could not afford. In Nairobi, a bakery owner crossed out tomorrow’s prices and wrote worse ones underneath.

The cost of a distributed problem is not the sum of its parts. It is the inability to prioritise.

Six weeks in, the conflict reaches a stalemate nobody wants to name.

In Tehran, a former lecturer named Arash stepped onto his roof and watched the city go dark in sections. Rolling blackouts. Fuel rationing. Half-working grids. Power had not collapsed. It had become local, which made it harder to see and harder to fix. Families ate dinner beside battery packs. The internet thinned to a trickle, then vanished. The state still had guns. It still had prisons. What it had lost was its monopoly on certainty. Everyone was improvising: the clerics, the Guard, the foreign powers, the street.

In Washington, the war room had its own shortage: sentences. The President wanted one he could live inside. No ground war. That line had held for six weeks. But language is a poor seawall. A Treasury official observed that markets could survive fear; what they hated was duration. Nobody at the table disagreed.

The resolution came from a quiet room in Muscat.

Oman, which has played this role for decades, carried the opening positions. Tehran’s ask: full sanctions relief on oil exports, a public commitment that regime change was not American policy, recognition of Iran’s right to enrich to 40%, and withdrawal of all new naval assets from the Gulf within 90 days. Washington’s ask: enrichment capped at 3.67% (the old JCPOA number), unlimited short-notice inspections of any site including military facilities, permanent dismantlement of the Fordow underground plant, and verifiable cessation of all proxy funding.

Both sides knew these were starting numbers. Neither could accept the other’s opening position. That is how negotiation works. The opening offer is not a proposal. It is a claim on the territory where the real conversation will happen.

The question was who would move first, and the answer, as in most negotiations, came down to time pressure. Washington was bleeding politically. Oil at $140 was a tax on every voter. A midterm election was eleven months away. Iran was bleeding economically, but the regime had decades of practice absorbing sanctions. It could wait. America’s clock was louder than Iran’s, and both sides knew it. Eighty per cent of concessions happen in the last 20% of available negotiating time. The side under more time pressure makes the bigger moves.

Over seven weeks of back-channel, the positions converged. The final settlement: enrichment capped at 20%, up from Washington’s ask of 3.67% but well below Iran’s demand for 40%. Inspections with 14-day notice on declared sites and 28-day notice on disputed sites, a compromise between unlimited access and the JCPOA’s delays. Fordow converted to a research facility, not dismantled but no longer enriching. Sanctions relief phased over 18 months, tied to verified compliance at each stage. No public statement on regime change, but a private letter through Oman. No naval withdrawal, but a drawdown to pre-conflict levels within six months.

Neither side got what it opened with. Both sides got enough to stop.

Three months after the first strike, the framework agreement was announced in Muscat. Hawks called it appeasement. Hardliners called it surrender. The publics in both countries barely registered it. A new crisis had already captured the headlines.

But the arithmetic held. The numbers balanced.

Wars end when both sides find a number they can live with. The number is never the one either side wanted.

The war began with cruise missiles over Natanz. It ended with a spreadsheet in Oman. Thirty-three kilometres of water decided more than thirty thousand tonnes of ordnance.

Frameworks

Find the chokepoint, not the target. The point of maximum leverage in any conflict, whether military or commercial or organisational, is rarely where the noise is loudest. It is the narrow passage everything else depends on. The Strait was worth more than every enrichment facility combined. In your world: what is the 33 kilometres that everything flows through?

Distributed problems defeat concentrated force. If someone can stretch you across four fronts simultaneously, your strength on any single front becomes irrelevant. The answer is not to fight harder everywhere. It is to refuse the frame and choose which front actually matters.

The side with the louder clock loses. Eighty per cent of concessions happen in the last 20% of available time. Washington had a midterm election. Iran had decades of practice waiting out sanctions. The side under more time pressure always makes the bigger concessions. In any negotiation, before you sit down, ask: whose clock is louder? If it is yours, either silence it or do not negotiate yet.

Get to the number faster. When both sides have proven they can hurt but cannot finish each other, the conversation will eventually become arithmetic. The people who cling longest to narrative and delay the numbers cause the most collateral damage. Reach for the spreadsheet before the speech.

The deal nobody celebrates is the deal that holds. If one side is delighted, the other is already planning to break it. The Muscat agreement left everyone mildly unhappy. Three years later, it was still standing.

This week I was going to publish the next framework in the regular sequence. Then I watched the news and realised the frameworks were already playing out in real time. So I wrote this instead.

Check out some of my other Frameworks on the Fast Frameworks Substack:

The Alpha Engine Framework for Venture & Angel Investing

Don’t Hire an Agency. Build a Memetic Engine - The memetic engine framework

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

BRITAIN’S GROWTH TRAP

Aditya Sehgal — Sun, 15 Mar 2026 03:29:23 GMT

THE GROWTH TRAP: A TLDR FOR PEOPLE WHO READ THE LAST PAGE FIRST

Billionaires are not the problem. They are the mechanism.

Britain has spent twenty-five years treating technology billionaires as a distributional failure. The data says the opposite.

Start with the number that should settle the argument. Strip technology out of both economies and compare American income per person with British income per person over the last twenty-four years. The structural gap has moved by roughly five hundred dollars. Per year, that rounds to nothing. The entire additional widening - roughly nine thousand dollars of extra annual income per person - came from one place. Not from harder-working Americans. Not from lower wages or weaker unions. From a single recycling loop that turned a £1.2 billion technology exit in 2002 into an ecosystem worth six trillion dollars today. Strip out the EU’s technology sector and the same flatness holds there too. The gap between all three economies and GDP per capita, once you remove technology, has barely moved in a generation.

That loop worked because of a specific mechanism. When PayPal was sold in 2002, the people who built it walked out with capital and hard-won operational knowledge simultaneously. They recycled both into the next generation of companies. Those founders recycled into the generation after. Each cycle started from a richer base than the last because the knowledge compounded alongside the capital. Britain watched every generation of that loop from the outside.

What Britain keeps losing is not abstract. It loses companies: DeepMind, sold to Google. ARM, sold to SoftBank, listed in New York. Graphcore, sold below its fundraising total. Autonomy, sold to Hewlett-Packard. Revolut, worth more than Barclays, being lobbied to consider London as a secondary option for a company it built here. Improbable, $3.4 billion, Cambridge. Herman Narula, Dubai, November 2025. The specifics differ. The direction has never changed. It also loses people: Britain had the biggest net drain of millionaires of any country in 2025.

The argument that Britain should have fewer of these people rests on a misreading of what founders actually keep. When Jeff Bezos stepped back as chief executive of Amazon he owned under ten percent of the company. More than thirty percentage points of what is now worth two and a half trillion dollars had been distributed outward - to pension holders, to staff through share awards, to the investors who funded every stage of growth. He built it. He kept a fraction. The same pattern holds everywhere. Founders keep a small slice of the equity. The equity itself is a fraction of total value created - wages, customer savings, tax revenues, and the ecosystem their exits seed all sit outside it entirely.

The argument that this is too much is an argument for generating less of what flows to everyone else.

Britain cannot redistribute its way out of a compounding gap.

This is what each chapter sets out to prove.

Chapter one: Britain has built a moral argument that treats wealth creators like criminals. It feels righteous. In 2025 it lost more millionaires than any other country on earth. The data shows this moral posture has been making ordinary people poorer for twenty-five years.

Chapter two: Strip out technology and the structural gap in GDP per capita between Britain, America, and Europe has not widened in twenty-five years. Every dollar of additional divergence is explained by a single recycling loop in which each generation of American founders seeded the next with capital and hard-won knowledge. Britain watched it being built. Then spent a generation stopping it from running here.

Chapter three: Britain has lost DeepMind, ARM, Graphcore, and Autonomy, and now risks losing Revolut. The definition of insanity is doing the same thing repeatedly and expecting a different result. Britain has been doing it for thirty years.

Chapter four: Britain has a five-year window to decide whether its first AI exits seed the next generation here or in California. The opportunity is real. The window is not permanent.

CHAPTER ONE

Britain’s Favourite Moral Argument Is Making Ordinary People Poorer

‘It is only fair that those with the broadest shoulders should bear the greatest burden’ is the most unchallengeable sentence in British politics. It has also done more damage to ordinary British living standards than any other idea of the last twenty-five years.

Fairness matters. But the specific policies Britain has built around this principle, applied with particular force to technology wealth, have been producing the precise opposite of fairness for a generation. This chapter is the proof.

The number

The number appeared in the summary. Here is what it means for the fairness argument. Strip technology from both economies and the gap in living standards between America and Britain has barely moved in twenty-four years. The entire additional widening - roughly nine thousand dollars of extra annual income per person - came from a single recycling loop that Britain watched being built in California and then spent two decades actively preventing from being replicated here. A policy framework that extracts from exits rather than compounding them has been choosing, year after year, to keep that gap exactly where it is. This is not a conservative argument or a libertarian one. It is what the data says.

What the culture produces

In America, when someone builds a company worth a billion dollars, the reaction of the average person and the average newspaper is roughly this: how did they do that, and how do I get there? The billionaire is evidence that the system works, a signal about what is possible.

In Britain, when the same thing happens, the reaction is roughly this: how much of that can we take? The billionaire is evidence that the system has failed, a problem to be redistributed, a story about luck and privilege and what someone owes.

One culture compounds. The other extracts. In Britain, founder wealth is treated as presumptively suspect. That suspicion is not just a feeling. It is encoded in policy that shifts with every budget, which breaks the long-horizon bets that compound loops require. The gap in the chart is what those two dynamics produce over twenty-five years. Both jaws of the trap must open simultaneously for the escape to work.

In 2025, Britain lost more millionaires than any other country on earth. Over sixteen thousand people with investable assets above a million pounds either left or were counted as departed. Britain is the only one of the world’s ten wealthiest countries whose millionaire population has shrunk over the past decade, while the United States grew its by seventy-eight percent.

Founders keep a fraction. Everyone else gets the rest.

Before the broadest shoulders argument continues to dominate British policy unchallenged, it is worth looking honestly at what founders actually keep when they build something of real scale. The numbers operate on two levels. Founders own a shrinking percentage of the equity they create - diluted by investor rounds, staff share schemes, and secondary sales. And equity itself captures only a fraction of total value created: wages, customer savings, supplier profits, tax revenues, and the ecosystem seeded by the exit all sit outside it. Put both layers together and the founder’s actual share of everything their company generates runs between one and three percent. Customer savings and spillovers are hard to measure precisely, but the direction is unambiguous: the equity slice is small relative to the total surplus a scaled company throws off. The rest belongs to everyone else.

In 2020, Jeff Bezos calculated where all the value Amazon created that year went. Employees received $91 billion in wages and benefits. Customers saved roughly $164 billion through lower prices. Third-party sellers made $25 billion in profit through the platform. Shareholders received $21 billion. Bezos personally received about one and a half percent of total value created. One and a half percent. When Amazon listed in 1997, he owned forty-three percent. When he stepped back as chief executive, he owned under ten percent. He built it. He kept a fraction.

Mike Lynch’s numbers tell the same story at a British scale. He built Autonomy, sold it for eleven billion dollars, kept approximately seven percent. The portion he kept he recycled into Darktrace and a generation of Cambridge technology companies. One founder, one exit, one reinvestment cycle, and the British AI ecosystem is measurably better for the companies it produced.

The argument that Britain should have fewer of these people is an argument that Britain should generate less of the ninety-seven to ninety-nine percent that goes to everyone else. It is not a fairness argument. It is a poverty argument dressed in the language of fairness.

What the policy produced

Founders who build companies to real scale face capital gains tax of 24% on gains above the first million pounds, with the preferential founder rate under Business Asset Disposal Relief having risen from 10% to 18% by April 2026, while economists continue to lobby for full alignment with income tax rates that would push it towards 45%. The direction is unmistakable even if the worst has not yet arrived. The policy framework has consistently chosen to extract revenue from the moment of wealth creation rather than keep that wealth inside the country where it can compound into the next company. In each case the founders paid their taxes, the state received its share, and the knowledge those founders had accumulated left the country along with the exit proceeds.

Britain is not being offered a choice between fairness and technology wealth. It is being offered a choice between two kinds of unfairness. The first is the unfairness of concentrated technology wealth inside Britain, addressable through progressive taxation on consumption and assets rather than on the act of building companies, through housing policy and public investment funded by a growing tax base. The second is the unfairness of a Britain that stays on the slow flat growth line while America compounds - a Britain in which the care worker, the teacher, the construction worker, fall further behind their American counterparts every year. Not because of anything a technology billionaire did to them, but because the growth that a domestic compound loop would have generated never happened here.

Britain has been choosing the second unfairness while believing it was choosing against the first.

The nine thousand dollars per person per year that Britain is not generating compared to America is not an abstraction. It is the nurse who cannot afford to buy a house in the town where she works. It is the NHS waiting list that a richer tax base could have shortened.

The second chapter explains how this recycling loop works and why Britain, by failing to build it, has been subsidising American prosperity with its own talent.

CHAPTER TWO

The Recycling Machine: Why Britain Is Paying for American Prosperity

In the year 2000, the gap in income per person between the United States and the United Kingdom was approximately fourteen thousand dollars. (All figures in this article use purchasing power parity adjusted dollars, which account for differences in the cost of living between countries. When a British figure appears in pounds it is noted explicitly. Otherwise assume dollars throughout.) In 2024 that gap is twenty-three thousand dollars. Nine thousand additional dollars of annual income per person, built up over a single generation, separating two countries that share a language, a legal tradition, broadly similar education systems, and outside of technology, broadly comparable economies.

Strip technology out of all three economies and recalculate. The structural gap in GDP per capita between the US, the UK, and the EU has barely moved in twenty-four years. The non-technology divergence is flat across all three. The technology gap is everything.

If American structural advantages were driving the divergence independently of technology, those advantages would be lifting American non-technology sectors above their British and European equivalents at a comparable rate. They are not. The structural gap has been stable. The growing gap is a different story. The PayPal alumni and their subsequent recycling cycles provide a specific explanation with unusual precision.

The two reasons America pulled away

The US technology sector outgrew every comparable economy for two reasons. The first is cultural: a society that celebrates wealth creation rather than resenting it produces more of it. In America, when someone becomes a billionaire, the average person’s reaction is: how did they do that, and how do I get there? In Britain, the reaction is: how much of that can we take? One culture compounds. The other extracts. In Britain, founder wealth is treated as presumptively suspect. That suspicion is encoded in policy that shifts with every budget, which breaks the long-horizon bets that compound loops require.

The second is mechanical: a recycling loop in which each generation of founders seeded the next with capital and hard-won knowledge, compounding value across four generations. The mechanical reason is the one Britain can copy without waiting for a cultural revolution.

The mechanism

In October 2002, eBay acquired PayPal for $1.5 billion. Approximately two hundred and twenty people walked out of that transaction with something that does not appear on a balance sheet: a library of hard-won knowledge accumulated through building a payments system at scale, surviving the dot-com crash, and learning what it actually feels like when a product grows faster than anyone expected.

Peter Thiel owned roughly 3.7 percent. Within two years he had invested five hundred thousand dollars of his exit in a twenty-year-old’s social network. He did not just bring money. He brought the specific knowledge library built at PayPal: what the early stages of a platform business actually require, how real identity works differently from avatar-based networking, when to resist investor pressure and when to accept it. His $500,000 became over a billion dollars. He used that to found Founders Fund, which invested in SpaceX, Airbnb, Palantir, Stripe, and dozens of others. Elon Musk took his share and built Tesla and SpaceX simultaneously. Reid Hoffman founded LinkedIn, then became one of the earliest and most consequential investors in the companies that followed.

None of this was charitable. It was commercially motivated recycling of capital and knowledge. The total effect of two hundred and twenty people doing versions of this simultaneously, across a decade, with each generation starting from a richer knowledge base than the last, is what produced the gap in the chart.

There is a name for the kind of knowledge that made this loop so powerful. A Xeme (more in my Xeme Framework) is the smallest unit of real-world learning, earned through doing something at scale and understanding from direct experience why it worked or failed. It is not what you learn from a lecture or a case study. It is what you carry after you have lived through something difficult at scale. When the PayPal group recycled their exits, they recycled both capital and xemes simultaneously. The Stripe founders who started Brex, Ramp, and Rippling began with Stripe’s full knowledge library available to them. The OpenAI alumni who founded Anthropic started with everything the preceding three generations had learned.

The key variable distinguishing the three economies is recycling density: the concentration of accumulated operational knowledge that stays inside a country’s borders, recycles freely into the next generation of domestic founders, and grows with each cycle. The United States has the highest recycling density of any technology ecosystem in the world. Britain’s recycling density in artificial intelligence is close to zero. Not because the knowledge does not exist here, but because the mechanism that would keep it here has not been allowed to develop.

Mario Draghi’s 2024 report on European competitiveness reached the same conclusion from a different angle. No European company worth more than a hundred billion euros has been founded from scratch in fifty years. Thirty percent of European startups valued above a billion dollars moved headquarters abroad between 2008 and 2021. In the five years to mid-2024, worker productivity per hour in the eurozone rose by 0.9 percent. In the United States it rose by 6.7 percent. Britain is running the same experiment as the EU with the same results.

The most technically serious objection is that American structural advantages - a single domestic market of three hundred and thirty million people, immigration designed to bring in exceptional talent, deeper capital markets - could plausibly explain the divergence independently. The structural advantages are real but they explain the existing gap, which has been stable for twenty-four years. They do not explain the nine thousand dollars of additional widening that opened up on top of it.

The third chapter counts the cost in specific companies and specific people, and asks whether Britain can still understand what it keeps losing before the next generation of exits follows the same path.

CHAPTER THREE

Britain’s Most Expensive Misunderstanding

Barney Hussey-Yeo founded Cleo in London in 2016. He was twenty-four years old, working from his bedroom in Shoreditch, building an AI-powered financial assistant for people who struggled to manage money. By 2018 he had raised serious capital. By 2020 he had moved the company entirely to the United States. Not because Cleo failed in Britain. Because the conditions that would have let it grow to the scale he had in mind did not exist here. In November 2024, standing at the sidelines of a fintech conference in London - still physically present, still a British founder - he told CNBC: “There’s so many founders already leaving, or already considering leaving - and they’re excited to go to Silicon Valley.” The excitement about Silicon Valley was not his. It was everyone around him.

He subsequently gave evidence to the House of Lords Communications and Digital Committee. What he said was not a complaint. It was a structural diagnosis. “There is now a ceiling on what the most ambitious founders can achieve here. A shortage of growth-stage capital, an onerous regulatory environment and a broken London Stock Exchange leave virtually no path to scale or list generational companies domestically.” He wrote in Fortune the same week: “Moving to America went beyond market size - it was a stark recognition that Britain stifles ambitious tech companies.”

Hussey-Yeo is the pattern.

The roll call

The specifics of each departure below differ. Some were acquisitions. Some were listings. Some were founders who left. The direction is unbroken.

In 2014, Google acquired DeepMind, founded in London, for approximately £400 million. The founders’ accumulated knowledge of how to build AI research at the frontier went into Google’s product organisation, where it has been compounding for American national income ever since. Britain received a campus and a headline.

In 2016, Microsoft acquired SwiftKey, whose machine learning technology sat on over three hundred million devices. The same year, Apple acquired VocalIQ out of Cambridge, voice AI that now lives inside Siri. Both generated the same pattern: British talent, British research, British operational knowledge, transferred to American balance sheets.

ARM is the wound that does not close. Founded in Cambridge, the company designed the chip architecture inside ninety-five percent of the world’s smartphones. In 2016, with no domestic capital source capable of funding its next phase of growth, the government approved its sale to SoftBank for £24 billion. Ministers called it proof that Britain remained open for business. Seven years later ARM chose Nasdaq, not London. The company that should have anchored a British semiconductor ecosystem trades in New York at a valuation above $150 billion.

Hermann Hauser co-founded Acorn Computers in Cambridge and spun out ARM in 1990. He had voted against the SoftBank sale. When the Nvidia acquisition bid came in 2020 he warned publicly that it would result in Britain having to ask the President of the United States for permission to use its own Cambridge-designed microprocessors. He described British politicians as “technologically illiterate” and the “root cause” of the problem. He has been saying this for thirty years. The company he built has changed hands twice. His argument has not changed once.

Graphcore should have been Britain’s Nvidia. Founded in Bristol in 2016, backed by Microsoft, Sequoia, and by Demis Hassabis as a personal investor, it raised $700 million and reached a $2.5 billion valuation. Then it ran out of runway because no domestic source of capital could fund the scale the AI hardware market required. In 2024 it was acquired for an estimated $600 million, below the total it had raised.

Stability AI ran out of money. The technology was real. The demand was real. The capital was not there.

Mike Lynch built Autonomy from a Cambridge laboratory into Britain’s largest software company and sold it to Hewlett-Packard for $11 billion in 2011. He put a portion into Invoke Capital, which was one of the first investors in Darktrace in 2013. One successful British founder, recycling one exit into many companies, exactly as the mechanism in the second chapter predicts. Then thirteen years of legal proceedings consumed him. He died in August 2024. The recycling stopped.

Darktrace is the proof that the mechanism works when conditions allow it to run. Lynch seeded it in 2013. It grew to employ over two thousand four hundred people, generated £625 million in annual revenue by 2024, and became one of the world’s leading AI cybersecurity platforms. One founder, one exit, one reinvestment cycle. The Lynch-Darktrace loop is exactly what this article is arguing for. It is also the clearest possible illustration of why the loop cannot depend on a single person. Lynch died and the compounding stopped. Darktrace has since been taken private by American firm Thoma Bravo. Britain needs a cohort, not a hero.

Revolut: built in London from 2015, sixty-five million customers, $4 billion in revenue in 2024, pre-tax profit of $1.4 billion, worth more than Barclays at $75 billion. Its chief executive has said publicly that the company prefers a US listing over London. The government is lobbying for a dual listing. The fact that lobbying is required tells you everything about the signal Britain has spent decades sending.

Herman Narula learnt to code at twelve, studied at Cambridge, co-founded Improbable at twenty-nine. The company became worth $3.4 billion. Before he left for Dubai in November 2025 he gave an interview to The National. “It’s less about the exit tax and more about not knowing what the next five budgets are going to hold and what random things are going to be placed inside them,” he said. “This is a country that was the birth of the Industrial Revolution, and now it doesn’t even have a single domestic company of note associated with AI.” He did not want to leave. He said so plainly: “I don’t particularly want to leave the UK - but I might want to one day and I don’t want to be banned from that option.”

What follows is not a list of companies that could not make it. It is a list of companies that did.

Born here. Grown elsewhere.

DeepMind. ARM. Graphcore. Autonomy. SwiftKey. Stability AI. Wise choosing New York. Revolut being lobbied to consider London as a secondary option. Improbable, $3.4 billion, Cambridge. Herman Narula, Dubai, November 2025.

Britain has been running this experiment for thirty years. The results are consistent.

The harder objection

The most uncomfortable challenge to this argument is the jobs one. AI is changing work across entire sectors. The disruption is coming whether or not Britain produces the companies driving it. A Britain with no domestic AI industry is not protected from those changes. It simply loses the tax revenues, the wages, and the new companies that a domestic industry generates alongside the disruption it causes. The founders who leave take the upside with them. The disruption arrives regardless.

The fourth chapter describes what the escape actually requires, and why the window for it is shorter than most people realise.

CHAPTER FOUR

The Escape

Picture two versions of Britain in 2040.

In the first, the AI companies being built in London and Cambridge and Bristol today produced their first significant exits between 2028 and 2033. Several of the founders who built them were persuaded to stay. Not by a summit or a press release, but by a specific set of signals that Britain sent between 2026 and 2030. Those founders recycled their exits into the next generation of British AI companies, bringing with them the hard-won knowledge of what it actually takes to build something at scale. By 2040 the loop has been running for roughly a decade. The ecosystem is not Silicon Valley. But it is self-sustaining, growing faster than every other part of the British economy, and closing the gap with America at a rate that would have seemed impossible in 2025.

In the second, the exits happened on the same schedule but the signals were never sent clearly enough, or were reversed by the budget that followed the one that introduced them. The founders who might have stayed did the arithmetic and concluded that a country whose rules shift with every political cycle was not worth building inside for thirty years. The loop never started. By 2040 Britain is still producing world-class research, still training exceptional engineers, still generating the ideas that other countries grow into industries. The gap with America, already twenty-three thousand dollars per person in 2024, is wider. Another report is commissioned to explain the failure. It reaches the same conclusions as the Draghi report of 2024, with an additional decade of evidence.

The difference is determined by the accumulation of decisions made between now and approximately 2030.

The trap has two jaws

The Growth Trap that holds Britain in slow linear growth has two jaws that work together, and both must open simultaneously for the escape to work.

The first jaw is policy. The combination of how capital gains are taxed on founder exits, how inheritance tax applies to operating businesses, how unstable the rules have become across successive budgets, and how little domestic capital exists at frontier scale, has made Britain resistant to the recycling loop. Not by design, but through the slow build-up of individually defensible decisions that nobody has assessed together for their combined effect.

The second jaw is culture. Britain has a story about wealth that it tells itself with great consistency: that concentrated wealth is evidence of a system that has failed, and the right response is to tax it and redistribute it. Those values are not the problem. The problem is applying them without distinguishing between wealth that stays and compounds inside Britain and wealth that simply leaves.

Change the policy without changing the culture and the culture finds ways to reverse the policy at the next election. Change the culture without changing the policy and the founders who are persuaded to stay still leave because the numbers do not work.

The signal

Herman Narula said explicitly before he left that the current rates were not the breaking point. The instability was. A country that changes its capital gains treatment for founders three times in a decade, that introduces and then removes relief for entrepreneurial assets, that threatens exit taxes and then retreats, is telling every founder something specific about the reliability of the environment in which they are being asked to make thirty-year bets.

Three things would change that signal. A statutory commitment, with cross-party support, that the core framework for technology founders will not change for ten years. A restoration of a genuinely preferential capital gains rate for founders who have built operating companies over many years - not traders, not passive investors - with a residency requirement so that gains stay inside the ecosystem. And the removal of inheritance tax from operating businesses where the founder is still actively building, because a forty percent tax levied at death catches the builders, not the inheritors.

In near-term revenue terms the cost of all three is close to nothing. The cost is political capital, which is why they have not happened.

The capital

Britain has one of the largest pools of long-term institutional capital in the world. Almost none of it flows into the British technology ecosystem at frontier scale. The average British pension fund is more likely to hold Apple, Microsoft, or Nvidia than the British AI company that might become their equivalent in thirty years.

The parallel requirement is a London Stock Exchange that is genuinely competitive with Nasdaq. ARM chose Nasdaq. Wise moved to New York. Revolut is being lobbied to consider London as a secondary option. The specific reasons - stamp duty on transactions, lower liquidity, slower index entry, lower valuations for growth companies - are all addressable. Every one of them is a political decision, not a technical problem.

The architecture

A data centre that takes four years to approve in Britain takes eighteen months elsewhere. Faster approval of the power generation and grid capacity that an AI economy requires is not a technology policy in itself. It is the condition on which technology policy can work at all. And a government that uses artificial intelligence to reform its own processes - faster planning decisions, faster visa processing, faster regulatory approvals - tells its founders something important: that it understands the technology seriously enough to stake its own credibility on it.

The twenty

The companies that could produce Britain’s founding cohort of AI recyclers are being built right now. Wayve is running Uber robotaxi trials in London in 2026. Isomorphic Labs, founded by Demis Hassabis and spun out of DeepMind, raised $600 million and is preparing to launch the first AI-designed drug trials. Nscale raised $1.1 billion to build AI infrastructure on renewable energy. Quantexa reached a $2.6 billion valuation. Synthesia reached a $2.1 billion valuation with hundreds of Fortune 500 companies as customers. Luminance is deploying proprietary legal AI across seventy countries. Cera delivers over sixty thousand AI-powered home healthcare appointments daily. CMR Surgical has raised over $600 million to build robotic surgery systems across Europe and Asia.

Each could produce a founder with the capital and operational knowledge to seed the next generation. Whether they do depends entirely on whether Britain has created the conditions that make staying preferable to leaving.

PayPal’s loop started with fewer than twenty recyclers. Britain needs enough density that a founder looking for capital and hard-won operational experience can find both without boarding a plane.

What the maths shows

Britain’s GDP is three trillion pounds. Growing at five percent annually for a decade means adding roughly one hundred and eleven billion pounds of new economic activity every year.

Twenty founders, each building a company worth a billion pounds and keeping seven percent, produce twenty companies worth fourteen billion pounds each. The total equity value created is two hundred and eighty billion pounds. But the wages those companies paid, the taxes they generated, the customer value they created, the ecosystem their exits seed - the total flowing outward to ordinary people is multiples of the equity figure alone. One generation of exits does not close a thirty-three-thousand-dollar gap. The loop does. Those exits seed the next generation of companies, which seed the generation after, each cycle starting from a larger base than the last. That compounding is what the five percent requires. Twenty founders gets you there for one year. The loop gets you there every year after that.

Five percent GDP growth requires not a debate about whether the wealthy deserve sympathy. It requires a decision about whether Britain would rather tax the exits it has or compound the exits it could have. The two are not the same choice. Britain has been treating them as if they were for twenty-five years.

The bill is a twenty-three-thousand-dollar gap per person per year. It is the nurse who cannot afford the house. It is the school underfunded by a Treasury that redistributed gains Britain did not generate.

The window

The PayPal exit was October 2002. Founders Fund opened in 2005. Three years separated the origin exit from the seeding of the second generation, and in those three years the architecture of the Silicon Valley compound loop was set.

Britain’s equivalent window is 2027 to 2032. Five years in which the decisions of a small number of people will determine the shape of the British technology economy for the following thirty years. The country that acts before the evidence is conclusive benefits from the compounding. The country that waits for the evidence commissions the report explaining why it fell behind.

Britain has been commissioning those reports for twenty-five years.

What will it actually take to make the next one stay?

Born Here, Grown Elsewhere examines Britain’s technology economy and the Growth Trap that prevents British-created value from compounding inside Britain. The Growth Trap framework identifies two simultaneous causes: policy choices that discourage founder reinvestment, and cultural beliefs that treat wealth creation as extraction rather than as national infrastructure. Both jaws must open simultaneously for the escape to work. The framework applies wherever compounding is blocked by simultaneous structural and cultural resistance - in national economies, in companies, in cities, and in the decisions of individuals.

A NOTE ON SOURCES AND METHOD

Income figures throughout come from World Bank and IMF purchasing power parity data. The technology sector attribution draws on ITIF research, BEA digital economy data, and LSE productivity analysis. The non-technology baseline figures are estimates derived from published data ranges rather than a single citable source; they are directionally consistent with ITIF and BEA research and should be read as approximations rather than precise calculations. Company valuations reflect figures available at the time of writing in March 2026. All GDP projections in Chapter Four are illustrative models built from published data ranges, not financial forecasts. All quotes attributed to named individuals are drawn from published interviews and public testimony cited in the bibliography below.

BIBLIOGRAPHY AND FURTHER READING

Income and GDP data

World Bank. GDP per capita, PPP (current international dollars). World Development Indicators. data.worldbank.org

International Monetary Fund. World Economic Outlook Database. October 2024 and April 2025 editions. imf.org

Office for National Statistics. GDP first quarterly estimate, UK: October to December 2025. February 2026. ons.gov.uk

Office for Budget Responsibility. Economic and fiscal outlook. March 2026. obr.uk

House of Commons Library. Gross domestic product (GDP): Economic indicators. March 2026. commonslibrary.parliament.uk

Technology sector attribution

Information Technology and Innovation Foundation. “Six Tech Industries Accounted for More Than One-Third of GDP Growth in the Last Decade.” May 2023. itif.org

Bureau of Economic Analysis. Measuring the Digital Economy. 2022. bea.gov

The Draghi Report and European competitiveness

Draghi, Mario. The Future of European Competitiveness. European Commission. September 2024. commission.europa.eu

Capital gains tax

House of Commons Library. Capital gains tax: recent developments. March 2026. commonslibrary.parliament.uk

HM Revenue and Customs. Capital Gains Tax: rates of tax. November 2024. gov.uk

The PayPal ecosystem

Soni, Jimmy. The Founders: The Story of PayPal and the Entrepreneurs Who Shaped Silicon Valley. Simon and Schuster. 2022.

SEC. “eBay Completes PayPal Acquisition.” October 2002. sec.gov

British technology losses and the millionaire exodus

TechCrunch. “SoftBank acquires UK AI chipmaker Graphcore.” July 2024. techcrunch.com

Wikipedia. “Arm Holdings.” Accessed March 2026.

Wikipedia. “Google DeepMind.” Accessed March 2026.

Henley and Partners. Private Wealth Migration Report 2025. henleyglobal.com

Witness testimony

Hauser, Hermann. Twitter, July 2016. twitter.com/hermannhauser

Hauser, Hermann. Interview with Cambridge Independent. August 2020. cambridgeindependent.co.uk

Hauser, Hermann. Interview with UK Tech News. June 2022. uktech.news

Hussey-Yeo, Barney. Evidence to House of Lords Communications and Digital Committee. November 2024. parliament.uk

Hussey-Yeo, Barney. Interview with CNBC. October 2024. cnbc.com

Hussey-Yeo, Barney. “UK tech’s problem isn’t taxes - it’s ambition.” Fortune. November 2024. fortune.com

Narula, Herman. “Tech entrepreneur Herman Narula: Why I’m swapping UK for UAE.” The National. November 2025. thenationalnews.com

Narula, Herman. Interview with The Telegraph. November 2025.

Revolut and British AI champions

Tech Nation. UK AI Sector Spotlight 2025. technation.io

The Tech Founders. “Top 100 UK AI Startup Investments in 2025.” thetechfounders.co.uk

Further reading

Thiel, Peter and Masters, Blake. Zero to One: Notes on Startups, or How to Build the Future. Crown Business. 2014.

Draghi, Mario. The Future of European Competitiveness: A Competitiveness Strategy for Europe. European Commission. September 2024.

Lacy, Sarah. Once You’re Lucky, Twice You’re Good: The Rebirth of Silicon Valley and the Rise of Web 2.0. Gotham Books. 2008.

Roelof Botha, various talks on the Sequoia model of patient, compounding capital. Sequoia Capital podcast archive and YouTube.

The Xeme Framework: Aditya Sehgal Xeme Framework

Check out some of my other Frameworks on the Fast Frameworks Substack:

The Alpha Engine Framework for Venture & Angel Investing

Don’t Hire an Agency. Build a Memetic Engine - The memetic engine framework

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

The Country That Got It Right

Aditya Sehgal — Sat, 14 Mar 2026 08:03:52 GMT

When US and Israeli strikes began hitting Iran on February 28th, the first question out of every analyst’s mouth was: what does this do to China?

China purchases roughly 90% of Iran’s oil exports. Nearly half of its seaborne crude transits the Strait of Hormuz. Its closest strategic partner was being bombed by its principal rival, and Beijing could do nothing about it. The conventional picture was a China caught badly exposed - diplomatically humiliated, energetically vulnerable, watching its supply lines burn.

The United States, meanwhile, had spent a decade becoming the world’s largest oil producer through the shale revolution. It had achieved energy independence. It was the country launching the strikes, not absorbing them.

Then the numbers came in. And everything flipped.

The Grey Rhino

In 2021, I wrote about four archetypes of risk. The one that makes sense of this war is the Grey Rhino.

You are on the African savannah. In the distance, a colossal rhino grazes. It is large, it is visible, you know exactly how fast it can run. It is not an ambush predator - it gives you every opportunity to move. The danger is not that the rhino is stealthy. The danger is that you get distracted, look at your phone, and by the time you look up again, it is already charging.

A Grey Rhino is a high-probability, high-impact threat that is entirely visible and almost entirely ignored. Not because it is hidden, but because acting on it requires paying a visible cost today for an invisible benefit tomorrow. That trade is psychologically brutal. Humans - and governments - are exceptionally bad at making it.

The Strait of Hormuz was a Grey Rhino. Every serious energy analyst knew that running a significant share of global oil supply through a 33-kilometre chokepoint between Iran and Oman was a structural vulnerability of the highest order. The geography was not going to change. The geopolitical logic of Iranian leverage over that chokepoint was not going to change. The question was always when, not whether, a crisis would arrive. And yet most countries continued running their economies through that single corridor year after year, because rerouting required investment that was expensive, slow, and politically unrewarding.

Two weeks into the war, we can see precisely who acted on the Grey Rhino and who did not.

The Bet That Did Not Look Like a Bet

Through the 2010s, China’s investment in renewable energy, electric vehicles, and grid infrastructure looked, from the outside, like industrial policy, pollution management, and climate signalling. It was all of those things. Underneath those rationales, it was also a deliberate response to a risk Chinese planners had mapped clearly: structural dependence on a chokepoint that, in any crisis, their adversary’s allies controlled.

Oil and gas supply 4% of China’s power mix. Across most major Asian economies, that figure runs between 40% and 50%. More than half of new cars sold in China are electric. Renewables met 80% of China’s new electricity demand growth in 2024. Oil shipments through the Strait of Hormuz account for approximately 6.6% of China’s total energy consumption. China entered this conflict holding 104 days of strategic crude reserves, rising to an estimated 140 to 180 days by year end.

The rhino charged. China was not standing in its path.

This is what acting on a Grey Rhino actually looks like. Not a single dramatic decision, but fifteen years of unglamorous infrastructure investment - each project individually justifiable on narrower grounds, collectively producing a resilience that only becomes fully visible when the crisis arrives. The cost was front-loaded and real. The benefit was back-loaded and invisible. Until now.

The benefit also turned out to be larger than the hedge alone. China did not just reduce its Hormuz exposure. It captured the manufacturing base for the global energy transition in the process. It now produces the dominant share of the world’s solar panels, lithium-ion batteries, and electric vehicle components, having invested a record $625 billion in renewable energy in 2024 - more than the rest of the world’s major economies combined. Every country now scrambling to reduce its own Hormuz exposure, which is most of them, will be buying Chinese technology to do it. The Grey Rhino response generated a Grey Rhino prize.

There was a third dividend nobody planned for. The air in Chinese cities is measurably cleaner than a decade ago. Coal power generation fell in 2025, the first annual decline in half a century, because renewable capacity expanded fast enough to absorb rising demand without it. One structural bet on energy independence delivered security, industrial dominance, and public health simultaneously. That is what happens when you act on the right risk early enough - the payoff compounds in directions you did not anticipate.

The Trap That Looked Like Dominance

The shale revolution was a genuine achievement. It made America energy independent in crude oil and gave Washington a geopolitical tool it had not previously held. It was also, by the logic of the risk framework, a response to the wrong risk.

Shale answered the question of whether America would run short of oil. It did not answer the question of whether oil itself was becoming a strategic liability. The Grey Rhino was not oil scarcity. It was oil dependency - the structural vulnerability of a global economy running on a fuel concentrated in unstable geographies, transited through military chokepoints, priced on markets that any regional conflict could detonate. The US, having achieved tactical energy independence, stopped looking at that rhino entirely.

The deeper error was statistical. Policymakers modelled Hormuz dependency as if the outcomes were normally distributed - occasional disruptions, manageable price spikes, diplomacy resolves it, back to normal. Gaussian thinking applied to a fat-tailed system. The actual distribution has a quiet, thin body - most years nothing catastrophic happens - and a tail that is not just fat but almost boundless. When a fat-tailed system fails, it does not fail a little. Currencies collapse, allies go into emergency rationing, oil crosses $100, a war has no exit. The downside is not worse than expected. It is a different order of magnitude from anything the model contained. Shale independence was so seductive precisely because it eliminated the thin-body risk - America’s own supply - while leaving full exposure to the tail. The US solved the small problem and declared the large one closed.

While China built the infrastructure for a post-oil economy, the United States cycled through energy policy with each administration. Obama invested in clean energy. The first Trump term reversed course. Biden reversed again. The second Trump term has reversed more aggressively than any predecessor, withdrawing clean energy incentives and pursuing fossil fuel dominance as explicit national strategy. No utility, grid operator, or manufacturer can make fifteen-year capital commitments in that environment. So they do not.

The manufacturing base for the energy transition - the batteries, the panels, the drivetrain components, the physical infrastructure of the post-oil economy - has migrated to China. The US won the current energy game and surrendered the structural position in the technologies that define the next one. It confused owning today’s market with owning tomorrow’s.

The cost of that confusion is now concrete. Oil above $100 damages the US economy through global price mechanisms even though no American crude transits Hormuz. The allies whose security commitments underpin US influence in the Indo-Pacific - Japan, South Korea, India - are absorbing the most acute pain from a conflict they had no part in starting. Every dollar those countries now spend accelerating their energy transitions will mostly flow to Chinese supply chains, because that is where the capacity lives. The US triggered the exact demand signal that cements China’s industrial advantage for the next decade.

The US went to the Middle East to project energy dominance. In doing so, it demonstrated what happens when you confuse tactical strength with structural resilience.

The League Table Nobody Wanted to Top

The war is generating a real-time ranking of which countries acted on their Grey Rhinos and which did not.

China sits at the top. Diplomatically embarrassed, commercially disrupted at the margins, but structurally insulated in ways no other major importer can match.

India sits in the middle, and the middle is genuinely uncomfortable. India imports more than 85% of its domestic oil needs. Half its crude transits Hormuz. It imports roughly 67% of its LPG requirements, and 90% of those shipments pass through the same chokepoint. The government has invoked emergency commodity legislation to ration cooking gas to 330 million households. Indian-flagged tankers slipped through the strait last week with their tracking transponders switched off to avoid Iranian targeting. New Delhi is negotiating with Tehran for permission to move its ships - asking a government it spent years distancing itself from under US pressure for safe passage to keep its restaurants open.

India’s renewable programme is real and accelerating - it dropped coal power generation in 2025 alongside China, the first simultaneous fall in half a century. But the direction of travel is not the same as the destination. India saw the Grey Rhino. It started moving. It has not moved fast enough, and this crisis is the invoice for that gap.

Japan and South Korea sit at the exposed end. Japan imports between 75% and 90% of its oil through Hormuz. South Korea sources roughly 60% of its crude the same way. Both entered this crisis with reserve levels measured in weeks. The Bank of Korea has activated a round-the-clock crisis task force. South Korea announced a $68 billion stabilisation fund. Japan is in a quieter panic. These are wealthy, technically sophisticated nations that watched the same Grey Rhino for decades and decided that reserves and diplomatic relationships were sufficient hedge. They were not.

Your Portfolio Has a Grey Rhino

Most investors manage risk tactically - diversifying across sectors, timing entries around current conditions, holding what performed well recently. That is Pesky Mosquito management: controlling individual exposures while the structural exposure underneath goes unexamined.

The Grey Rhino question for an investor is different. Not: what is the biggest risk in my portfolio right now? But: what structural shift is highly probable, high impact, and being widely ignored - and am I positioned on the right side of it before it arrives?

The clean energy transition was a visible Grey Rhino in 2021 when I wrote the original framework. It is a more urgent one now. A major military conflict has just demonstrated why every country on earth has urgent reason to reduce dependence on oil running through a single chokepoint. The policy response to this crisis will accelerate clean energy deployment in ways that voluntary climate commitments never could. The question for investors is not whether that deployment happens. It is whether you are positioned in the companies and geographies that will supply it.

The second Grey Rhino is the technology infrastructure question. The country or company that manufactures the components of the next energy system occupies the same structural position that oil exporters held in the twentieth century. China understood that and invested accordingly. Most Western investors are still treating Chinese clean energy dominance as a trade policy irritant rather than a structural reality about who owns the supply chains of the next fifty years. That is not a political position. It is a portfolio miscalculation.

The Rhino in Your Boardroom

The competitive version of this story is one most business leaders recognise immediately and act on almost never.

Large companies treat startups as Pesky Mosquitoes - individually irritating, manageable in isolation, unlikely to cause real damage. For a while, that assessment is correct. The startup is small, its product is rough, its market share is a rounding error. The mosquito is not worth the distraction.

The error is categorical, not factual. The dangerous startup is not a Pesky Mosquito at all. It is a Grey Rhino wearing Mosquito clothing.

Kodak invented the digital camera in 1975. It filed the patent, examined what digital photography would do to its film business, and put the technology in a drawer. The rhino was not just visible - it was sitting in their own laboratory. They chose the current market over the next one and paid for it with the company.

The structural pattern is always the same. The incumbent optimises for the current competitive environment. The challenger optimises for the next one. For years, optimising for the current environment looks correct - the numbers are real, the customers are real, the margins are real. Then the environment shifts, and the challenger’s long investment in a capability that looked unnecessary becomes the only capability that matters.

The Grey Rhino question for a business is not: who is our biggest competitor today? It is: what structural shift in our market is highly probable, high impact, and being treated as someone else’s problem - and is there a company, perhaps still small, that has been quietly building for that shift while we defend our current position?

There is almost always a company quietly building for the shift you have decided is someone else’s problem. The question is not whether it exists. It is how much of the 33 kilometres it has already closed.

The Rhino Is Already Charging

The country everyone assumed was screwed built the hedge. The country everyone assumed had it figured out built the trap.

The Strait of Hormuz will eventually reopen. Oil prices will stabilise. This war will end. What will not revert is the lesson this crisis is teaching, at enormous cost, to every government, investor, and business leader paying attention: structural resilience is built before the crisis, or it is not built at all.

Grey Rhinos do not disappear because you stop watching them. They just get closer.

The question worth sitting with today is not what your Grey Rhino is. You already know. The question is whether you are looking at your phone.

Sources

China’s energy position and Hormuz exposure

CNBC, 9 March 2026: China oil shock, Iran war, Hormuz energy transition - 4% power mix figure, EV penetration, 80% renewables share of new demand, 6.6% total energy consumption through Hormuz, 104-day reserves, OCBC analyst quotes
CleanTechnica, 13 March 2026: How China Is Avoiding the Straits of Hormuz Curse - coal power decline in 2025, economy redesigned around electricity
CNBC, 3 March 2026: Strait of Hormuz closure - which countries most impacted - 40% seaborne oil through Hormuz, LNG exposure data
OilPrice.com, January 2026: Five Key Trends Driving China’s 2026 Energy Strategy - $625 billion renewable investment figure

China’s geopolitical position

Middle East Institute, 5 March 2026: China in the Crossfire - Belt and Road Gulf exposure, strikes beginning February 28
Chatham House, February 2026: China is Playing the Long Game over Iran - China’s historical military non-intervention
CNN, 4 March 2026: Why has Beijing done very little? - 90% of Iran’s oil exports to China, rare earth export ban
Foreign Policy, 10 March 2026: US May Need China to Help Contain Iran War - China purchases 90% of Iran’s oil, one-third of Iran’s total trade
Chinese Foreign Ministry, 2 March 2026: Spokesperson Mao Ning Press Conference - official Chinese position on Hormuz

India

Bloomberg Opinion, 12 March 2026: Iran War Exposes India’s Vulnerability - LPG shortage, cooking gas crisis
Business Standard, 12 March 2026: India in talks with Iran for safe passage - transponders switched off, 28 vessels stranded, non-Hormuz sourcing rising to 70%, LPG production up 28%
CNBC, 2 March 2026: India hit by high oil prices - 85% import dependency, half of crude through Hormuz
CNBC, 10 March 2026: Iran war threatens LPG supply - 67% LPG imported, 90% of those through Hormuz
The Week India, 11 March 2026: The Hormuz Flashpoint - 2.5 to 2.7 million barrels per day through Hormuz, 330 million households on LPG

Japan and South Korea

Asia Energy Crisis analysis, March 2026: Asia Energy Crisis 2026 - Japan 75-90% Hormuz exposure, South Korea 60%, $68.3 billion stabilisation fund
Green Central Banking, 9 March 2026: Clean energy, not LNG, is Asia’s best hedge - Bank of Korea crisis task force, Singapore and Bangladesh exposure

Coal decline

Carbon Brief, January 2026, via CleanTechnica: China and India coal power generation fell simultaneously in 2025, first time in half a century

The Risk Framework

Aditya Sehgal, December 2021: The Risk Management Framework - Grey Rhino, Black Swan, Pesky Mosquito, Rare Panda archetypes

Check out some of my other Frameworks on the Fast Frameworks Substack:

The Alpha Engine Framework for Venture & Angel Investing

Don’t Hire an Agency. Build a Memetic Engine - The memetic engine framework

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

ARC — Audience × Resonance × Craft

The ARC Framework + ARC GPT

Aditya Sehgal — Fri, 13 Mar 2026 22:02:03 GMT

Try ARC (free Custom GPT): I built a GPT version of this framework so you can diagnose an audience, pick an arc, and draft the skeleton faster. It’s free for readers here, and your data is Private. Access it here: ARC — Audience × Resonance × Craft

Every founder I work with has the same problem. They know their numbers. They believe in their product. They have rehearsed the pitch. And they cannot get the room to care.

The data is right. The conviction is real. The story is wrong. Not wrong in a grammatical or stylistic sense. Wrong in a structural sense - the kind of wrong that no amount of rewriting will fix because the problem is not the words. It is the architecture underneath them.

The Model

ARC = Audience x Resonance x Craft

The relationship is multiplicative. If any layer is zero, the whole thing is zero. Beautiful writing aimed at the wrong audience fails. The right audience hearing the wrong arc shape feels manipulated or unconvinced. And the perfect arc told in the wrong format falls flat.

The Spine: Four Steps

If you remember nothing else, remember this:

1. Diagnose the audience. “They currently believe , and I need them to believe .”

2. Pick the arc. Match their state to one of seven shapes. (See the Fast Picker below.)

3. Build the skeleton. HOOK - TENSION - TURN - RESOLUTION - SHIFT. Start at the Turn.

4. Tune the craft. Match tone, evidence type, and format grammar to the medium.

That is the whole method. Everything that follows is reference material for each step - go deep where you need to, skip what you already know.

The Cereal Box That Built a $75 Billion Company

In 2008, Brian Chesky and his co-founders pitched Airbnb to fifteen of Silicon Valley’s top venture capitalists. They had a working product, paying customers, and a clear market. Every single investor said no. The idea that strangers would pay to sleep in other strangers’ homes sounded absurd. The data was not enough. The deck was not enough. The logic was not enough.

Rejected by VCs, the founders turned to a stunt to survive. They designed custom cereal boxes - “Obama O’s” and “Cap’n McCains” - themed around the 2008 presidential election, and sold them for $40 each. They moved over a thousand boxes and made $30,000. Enough to keep the lights on. Nothing more.

Then they walked into Paul Graham’s office at Y Combinator. The pitch went the way the others had gone - Graham was sceptical. People living with strangers? Before he could leave, Chesky handed him a cereal box. Graham turned it over. Then he said something that changed the company’s trajectory: “If you can convince people to pay $40 for a $4 box of cereal, maybe you can convince strangers to live with each other.”

Y Combinator invested $20,000 for 6% of the company. That stake became worth over $4 billion.

The cereal box was not a business plan. It was a story - a physical object that proved something no slide deck could: these founders were resourceful enough, creative enough, and relentless enough to figure it out. The cereal box did not explain Airbnb. It made Paul Graham believe in the people building it. Fred Wilson, the VC who passed on Airbnb, kept a cereal box on his conference room credenza for years afterward as a reminder of what he missed.

Chesky’s first fifteen pitches had the right data. The cereal box had the right arc. In ARC terms: those pitches were strong on Craft (a good deck) but wrong on Resonance (a logical argument to an audience that needed to feel something visceral about the founders). The cereal box fixed the Resonance layer. Same company, same data, different arc.

Why Smart People Fail at Storytelling

I have sat through hundreds of pitches, brand presentations, and sales decks. The people giving them are almost always smart, prepared, and passionate. The stories they tell are almost always forgettable. Not because the people lack talent. Because they lack a diagnosis.

Most business stories fail for one of three reasons. Once you can name the pattern, you can fix it in an afternoon.

The Data Dump. The storyteller skips straight to a spreadsheet. The information is correct, comprehensive, and well-formatted. It is also forgettable. This is Chesky’s first fifteen pitches - strong data, no arc. They had every number right and still could not move a room.

The Empathy Trap. The storyteller understands who they are talking to and articulates the problem perfectly - but then never goes anywhere. They mirror the audience’s frustration without offering a turn. The audience feels understood and then... nothing. No new information, no reframe, no proof, no call to action. This is common in brand content that is “relatable” but forgettable. Empathy without a pivot is sympathy. Sympathy does not move people.

The One-Arc Founder. The storyteller has one story - usually an origin story - and tells it in every context. To investors, to customers, to employees, to media. The origin story is powerful in its right context. It is wrong for a sales conversation, wrong for a fundraise, and wrong for recruiting. Different audiences in different moments need different shapes.

I learned how expensive this last one is the hard way. A few years ago, I was advising a founder on her Series A pitch. She had a genuinely moving origin story - the kind that makes you lean in. I helped her build the entire deck around it. Origin arc, the full structure. The VCs listened politely and passed. Every single one. The feedback, when it came, was consistent: “Great founder. Not sure about the market.”

We had told the wrong story. These investors were not wondering about the founder. They were asking whether the market was big enough and moving fast enough. They needed a completely different story shape - one that said the game has changed, here is the structural gap, this founder is positioned to own it. We rebuilt the deck. Same company, same founder, same data. Different arc. She closed her round in three weeks. The origin story became a thirty-second aside in the founder slide. It was still beautiful. It was just no longer the load-bearing wall.

The question that kept nagging me after that: why do we keep getting this wrong? Why do smart people with the right data, the right product, and the right conviction keep failing to move the room?

The answer, I eventually realised, is that we treat storytelling as writing. As a creative act. Something you are either good at or you are not. But it is not writing. It is diagnosis. And diagnosis has a method.

Layer 1: Audience - The Story Before Your Story

A doctor who prescribes without examining the patient is negligent. A storyteller who writes without understanding the audience is guessing. The Audience layer is the examination.

Every person you are trying to reach is already mid-film. They have a plot running in their head - about their market, their career, their company, their problem. That plot has characters, a narrative, and an emotional register. Your story does not start on a blank page. It starts on a page that is already half-written. If you ignore what is already written, your story feels irrelevant. If you acknowledge it, your story feels like it was meant for them.

The Audience layer answers three questions:

What is their current narrative? An investor who believes AI is overhyped lives in a different story than one who believes AI is underinvested. A customer who thinks their current solution is “good enough” occupies different territory than one who is actively frustrated. Paul Graham’s narrative was: “This idea sounds crazy, and most crazy ideas fail.” Chesky could not win by arguing against that narrative. He had to sidestep it entirely - from “is this idea logical?” to “are these founders unstoppable?”

What frame are they already in? Every interaction has a psychological frame the audience brings with them. Oren Klaff identifies four: the Power frame (”convince me”), the Analyst frame (”show me the data”), the Time frame (”I have five minutes”), and the Intrigue frame (”tell me more”). If your audience is in an Analyst frame and you lead with emotion, it feels manipulative. If they are in an Intrigue frame and you lead with a spreadsheet, you kill the curiosity. This is their incoming frame - the state you are diagnosing. (What frame you set in response is a separate move. That is Frame Control, covered later.)

What is the desired shift? Every effective business story moves the audience from one belief to another. The shift must be specific enough to complete one sentence: “I want them to move from ____ to _____ . If either blank takes more than fifteen words, the shift is not specific enough. “I want them to be interested” is not a shift. “I want them to move from ‘this idea sounds crazy’ to ‘these founders will figure it out’” is a shift. “I want them to move from ‘we are fine’ to ‘we are exposed’” is a shift. If you cannot name it, you do not have a story. You have content.

In my Trust Framework, I describe trust as a fraction: credibility, reliability, and intimacy in the numerator, self-interest in the denominator. The desired shift often maps directly to one of those variables. A pitch deck shifts credibility. A customer story shifts reliability. A Belonging narrative shifts intimacy. And a story that feels manipulative - one where the audience senses the storyteller’s self-interest - collapses all three, because the denominator just got very large.

Before you write anything, complete this card. If you cannot fill every line, you are not ready to write. Go back to the beginning of this layer.

ARC Diagnosis Card

Audience: They currently believe:

Their frame: Power / Analyst / Time / Intrigue

Desired shift: from to

Arc: Skeleton:

Hook: ___
Tension: ___
Turn: ___
Resolution: ___
Shift: ___

Craft: Format / Length / Evidence type / CTA

One line per beat. If you cannot complete the Turn line in one sentence, the story does not have a hinge yet.

Layer 2: Resonance - The Seven Arc Shapes

A musician does not play random notes. She chooses a key. The key determines which notes create harmony and which create dissonance. The arc shape is the key of your story. It determines which emotional and strategic notes will resonate with your audience and which will feel wrong.

There are seven arc shapes, organised into three pairs and a standalone.

The Fast Picker

They need to know why you exist - use Origin. They need to feel the ground shifting - use Disruption. They need to see proof it works - use Transformation. They need to feel the clock ticking - use Tension. They need to feel they belong - use Belonging. They need to hear what you learned the hard way - use Insight. They need to see where you are going - use Vision.

When two arcs seem close: If they do not yet agree the world changed, lead with Disruption. If they agree but are not moving, lead with Tension. If they want evidence before they commit, lead with Transformation. If they are asking “why you?”, lead with Origin or Insight depending on whether the answer is your story or your scar.

Build stories tell the audience where you came from and where you are going: Origin (why this exists) and Vision(where we are building toward).

Proof stories show the audience what changed and what you learned: Transformation (what changed for someone) and Insight (what I learned the hard way).

Action stories create urgency and demand a response: Disruption (the game has changed) and Tension (the clock is ticking).

Belonging stands alone as the identity arc - it does not tell a story about an event. It tells the audience who they are.

Choosing the right one is not a creative decision. It is a diagnostic one. Match the arc to the audience’s narrative state and the desired shift you diagnosed in Layer 1, and the story almost writes itself. Get the match wrong and no amount of beautiful writing will save it.

Resonance has two checks, not one. The first is arc selection - did you pick the right shape for this audience and this shift? The second is Turn credibility - does the pivot moment earn its weight? You can pick the right arc and still fail Resonance if the Turn is weak. A Disruption arc where the “shift” is vague, or a Transformation arc where the “catalyst” is unconvincing, has the right shape but no hinge. When a story feels hollow despite being structurally correct, the Turn is almost always where it broke.

Arc 1: Origin - “Why This Exists”

Dramatic engine: Dissatisfaction → Discovery → Conviction

Sara Blakely spent two years selling fax machines door to door. Every morning she put on trousers that did not fit right. She cut the feet off her pantyhose. That solved the problem, temporarily, badly. The frustration accumulated - not a eureka moment, but a slow boil. Two years of cutting pantyhose before she filed a patent. That is Spanx’s origin story. Not “visionary entrepreneur builds a billion-dollar company.” It is “a woman who was annoyed every single morning finally did something about it.”

The Origin arc tells the story of why something had to exist. The best origin stories do not start with the founder. They start with the frustration - a moment where the world revealed a gap that could not be ignored. The discovery is not a lightning bolt. It is a slow accumulation of evidence that the frustration is not personal but universal. And the conviction is the point where walking away became harder than building. That is where the audience leans in - not because the solution is clever, but because the pain is recognisable. The product is the proof. The problem is the story.

The frame this sets is Moral Authority. The storyteller earns credibility through the authenticity of the dissatisfaction, not through credentials.

Best for: Brand narratives, founder stories, “about us” pages, brand films, company culture documents, media interviews.

Example compressed: “Every performance review I ran as a manager felt like theatre. The employee knew the score before I said it. I knew the conversation would change nothing. After twelve years of running these rituals, I stopped asking how to make reviews better and started asking why we do them at all. That question became [company].”

Arc 2: Disruption - “The Game Has Changed”

Dramatic engine: Old World → Shift → New World → Who Wins

“In 2019, Indian D2C brands competed on price. In 2025, they compete on community. The brands that built audiences own their economics. The brands that bought traffic rent theirs. The game changed. Most have not caught up.”

That is a Disruption arc in six sentences. It does not mention a product. It does not mention a company. It names a shift in the world and forces the audience to ask: which side am I on?

In 2007, Nokia had 50% global market share in mobile phones. By 2013 it had 3%. Nokia did not make bad phones. Nokia made phones for a world that no longer existed. The shift was not “smartphones arrived.” The shift was that the phone stopped being a communication device and became a computing platform. Every company that understood the shift thrived. Every company that saw it as incremental was consumed by it. In my Exponential Growth Framework, I describe the S-curve that governs this: lag phase, log phase, plateau, decline. Nokia was on the decline curve of one S and refused to jump to the next. The Disruption arc names that jump.

The Disruption arc is not about your company. It is about a change in the world that makes your company necessary. The dramatic engine runs on a single claim: the rules have changed, and most people have not noticed yet. Andy Raskin’s strategic narrative framework - the most influential sales deck structure of the last decade - lives entirely within this arc. His insight was that the best sales decks lead with market change, not product features.

The natural frame here is Prize. You are not asking for attention. You are revealing something the audience needs to know. You are not being evaluated. You are selecting who gets to participate.

Use it for pitch decks, investor narratives, category creation, thought leadership, product launches, keynotes, and fundraising. Anywhere you need the audience to feel the ground shifting under their feet.

Arc 3: Transformation - “What Changed”

Dramatic engine: Stuck → Struggle → Catalyst → New Reality

Kurt Vonnegut once drew the shapes of stories on a blackboard. His most universal shape was “Man in a Hole”: somebody gets into trouble, then gets out of it. He drew a line that dipped below the axis and climbed back above it. That shape - the dip and the climb - is the Transformation arc. The dip is essential. Without it, the story is a testimonial (”we loved the product”). With it, the story is a narrative (”we were drowning, we found this, and everything shifted”).

The Transformation arc is proof. It tells the story of someone who was in the audience’s situation and came out the other side. It does not ask the audience to believe a claim. It asks them to witness a journey. In my Pretotype Framework, I describe how to test whether something works before committing to building it. The Transformation arc is the narrative version of a pretotype - it lets your audience test the outcome in their imagination before committing to the purchase. The Struggle is not a minor inconvenience - it is a genuine obstacle with consequences. The Catalyst is what changed, but the emphasis is on the change, not the catalyst. And the New Reality is specific and measurable.

The frame is Intrigue. The audience stays because they need to know what happened next. The sale happens in the audience’s own mind, not in the narrative.

Best for: Case studies, customer stories, testimonials, sales conversations, demo narratives, before/after content.

Example compressed: “Priya’s supplement brand was spending 40% of revenue on paid ads and breaking even. Six months later, her repeat purchase rate hit 58% and her CAC dropped by half. What changed was not her product.”

Arc 4: Tension - “What Is At Stake”

Dramatic engine: Stakes → Escalation → Fork in the Road → Resolution Path

In 1962, JFK did not say “we should probably go to the moon at some point.” He said: “We choose to go to the moon in this decade.” The deadline was the story. Not the destination, not the technology, not the budget - the clock. That is the Tension arc. It creates urgency. It makes people act rather than consider.

Where the Disruption arc says “the game has changed” (intellectual urgency - I need to understand), the Tension arc says “the clock is ticking” (emotional urgency - I need to act now). The Escalation reveals the stakes are higher than initially understood. The Fork in the Road presents the choice: act or accept the consequences. And the Resolution Path shows that action is possible and achievable - but only if taken now.

This arc draws on PAS (Problem-Agitate-Solve) but adds the Fork. PAS moves linearly from problem to solution. The Tension arc forces a choice, which is psychologically more activating than a recommendation. The audience feels not just that they have a problem, but that waiting has a cost.

The frame is Time/Scarcity. In my Risk Framework, I sort risks into four buckets: Grey Rhinos (high probability, high impact), Black Swans (low probability, high impact), Pesky Mosquitoes, and Rare Pandas. The Tension arc works best when the stakes are a Grey Rhino - obvious, high-impact, and approaching fast. The failure mode is manufacturing false urgency around a Pesky Mosquito, which destroys trust permanently. When the stakes are real, clarity about deadlines is not manipulation. It is a service.

Best for: Sales narratives, fundraising, calls to action, campaign launches, competitive displacement, crisis communication.

Example compressed: “Your three largest competitors migrated to headless commerce last quarter. Your site speed is 4.2 seconds. Theirs average 1.1. Every month you wait, you lose an estimated 12% of mobile conversions to them. The migration takes six weeks. The question is not whether to move. It is whether to move before or after peak season.”

Arc 5: Belonging - “Who We Are Together”

Dramatic engine: Isolation → Recognition → Tribe → Mission

In 1997, Apple was ninety days from bankruptcy. Steve Jobs did not launch a product campaign. He launched “Think Different.” The ad did not mention a single Apple product. It mentioned Einstein, Gandhi, Picasso, Lennon. The message was not “buy our computer.” It was “you are the kind of person who sees the world differently, and so are we.” Within two years, Apple’s stock had quadrupled. A membership card, not a sales pitch.

The Belonging arc does not sell a product. It sells an identity. Isolation is the feeling that you are the only person who sees this problem. Recognition is discovering others who share your worldview. Tribe is the group identity. Mission is the collective purpose that gives the tribe meaning beyond membership.

People do not comparison-shop their identity. They commit to it. In my Identity Migration Framework, I describe how professional identity climbs four rungs: from task to responsibility to capability to purpose. The Belonging arc speaks directly to purpose-level identity - it asks people to recognise themselves in a mission, not a product. This is why Belonging stories are selective by design - they deliberately exclude to strengthen the bond with those who remain. Patagonia did not sell jackets. It sold membership in a tribe that values the planet over profit. The product is a badge, not the point.

The frame is Identity. The audience is not being asked to buy. They are being asked to recognise themselves. Are you one of us?

Best for: Community building, brand identity, internal culture, employer branding, movement creation, loyalty programmes, recruiting.

Example compressed: “We do not hire people who want to work at a startup. We hire people who cannot stop thinking about why Indian supply chains are broken and believe they can fix it. If that sentence made you sit up, keep reading.”

Arc 6: Insight - “What I Learned the Hard Way”

Dramatic engine: Assumption → Experience → Rupture → Reframe

Howard Schultz visited Milan in 1983 and saw something that did not exist in America: a neighbourhood coffee bar where people lingered, talked, and treated coffee as an experience, not a commodity. He pitched the idea to Starbucks’ founders. They said no. The insight that coffee could be a “third place” was born not from market research but from standing in an Italian piazza watching people do something Americans did not know they wanted. The assumption (”coffee is a grocery product”) broke against experience. The reframe (”coffee is a place”) became a $100 billion company.

The Insight arc is the engine of thought leadership and personal brand. It requires vulnerability about what you got wrong. In my Xeme and Learning Framework, I call the smallest unit of real-world learning a xeme - something earned through testing, not reading. The Insight arc is a xeme told as a story. The audience trusts it precisely because it was paid for in experience, not borrowed from a textbook. Matthew Dicks calls the pivot the “five-second moment” - the instant understanding shifts. In the Insight arc, that moment is the Rupture. Everything else exists to set it up and pay it off.

The storyteller is a fellow traveller, not an authority dispensing wisdom from above. A CEO sharing a failure teaches more than a CEO sharing a success, because the failure demonstrates judgment, not just outcomes. This is why the best LinkedIn content follows this pattern, and why the worst states principles without showing the price of learning them.

The frame is Intellectual Authority - but earned, not claimed. Credibility comes from the quality of the experience and the honesty of the admission.

Best for: LinkedIn posts, thought leadership, keynote speeches, podcast appearances, personal branding, blog posts, advisory contexts.

Example compressed: “For ten years I hired for skills. The best developer, the best marketer, the best operator. My teams were talented and miserable. Then I lost three A-players in one quarter and finally asked the question I had been avoiding: what if the problem is not them? It was not. The problem was that I built teams for output, not for trust. I now hire for one thing first: do I trust this person to tell me what I do not want to hear? Everything else can be taught.”

Arc 7: Vision - “Where We Are Building Toward”

Dramatic engine: Present Limitation → Imagined Future → The Path → The Invitation

In February 2014, Satya Nadella became CEO of a Microsoft that Wall Street had written off. Windows was declining, mobile had been lost to Apple and Google, and the company’s identity was anchored to a product era that was ending. Nadella’s first company-wide memo did not defend what Microsoft had been. It described what the world was becoming: “We live in a mobile-first and cloud-first world.” Six words that reframed a 39-year-old company. He did not have proof yet. Azure was small. Office 365 was early. But the vision was specific enough to be credible and ambitious enough to demand a choice - join or watch. Within a decade, Microsoft’s market capitalisation went from $300 billion to over $3 trillion.

The Vision arc tells the story of a future that does not exist yet and makes the audience want to build it. Where the Disruption arc says “the game has changed,” the Vision arc says “here is the game we are creating.” Disruption is reactive. Vision is generative - it brings something new into existence through the force of the narrative itself.

Present Limitation is an honest acknowledgement of what is broken. The Imagined Future is a vivid, concrete picture of what becomes possible. The Path is the credible route from here to there. And The Invitation is the call to participate - not as a customer, but as a builder. This arc sits on the knife-edge between inspiration and delusion. The difference is specificity. “We are going to change the world” is delusion. “We live in a mobile-first and cloud-first world” is vision - specific, directional, and testable.

The frame is Prize/Invitation. This is happening whether or not you join. But it will be better if you do.

Use it for mission statements, brand manifestos, internal rallying narratives, recruiting, long-term investor communication, and keynotes where you are defining a category rather than competing in one.

Example compressed: “In three years, the phrase ‘product returns’ will not mean sending something back. It will mean the product returns to the supply chain, is remanufactured, and ships again. We are building the infrastructure for that world. The question is not whether it happens. It is who builds it.”

The Universal Skeleton

Here is the part that makes the whole thing practical. Every arc shape, regardless of type, runs on the same five-beat skeleton. I call it the accordion because the same structure compresses into a tweet and expands into a keynote. The beats are:

HOOK - The opening frame. Its only job is to stop the scroll, hold the gaze, or open the ear. Chesky’s Hook was a cereal box. Schultz’s Hook was an Italian piazza. A Hook works by creating a gap - between what the audience expects and what you deliver. “Revenue grew 30%” is not a hook. “We almost shut down in March” is. The gap creates a question, and the question creates attention.

TENSION - The dramatic problem. Without Tension, a story is an announcement. Tension comes from conflict, contradiction, stakes, uncertainty, or unresolved questions. The audience stays because they need to know how it resolves. In the Nokia example: “50% market share to 3% - what happened?” In the Spanx example: “Two years of cutting pantyhose - what did she do?” The moment Tension disappears, attention disappears with it.

TURN - The pivot moment. The non-negotiable core. The hinge that everything before builds toward and everything after pays off. In the Insight arc, the Turn is the Rupture. In the Transformation arc, the Turn is the Catalyst. In the Disruption arc, the Turn is the Shift. It occupies the smallest proportion of time or space but carries the greatest weight. It never gets cut, even in the shortest format.

RESOLUTION - What happened, what is possible, what changed. The Resolution pays off the Tension. It must be specific. “Things got better” is not a Resolution. “Repeat purchase rate went from 12% to 58% in six months” is. Chesky’s Resolution was Y Combinator’s investment - but it landed because the cereal box had already made Graham believe.

SHIFT - What the audience should now believe, feel, or do differently. The strategic delivery. The frame the audience leaves with. If the Shift lands, the story worked. If it does not, nothing else matters. The Shift connects back to the Audience layer: it is the desired change you diagnosed before you started writing.

The Accordion: Scaling the Skeleton

In longer formats, the Tension beat is built from two to four sub-stories that each escalate the stakes. The Resolution beat is built from two to three proof points that each compound the evidence. The Turn stays singular - one moment, one hinge.

Frame Control: The Frame You Set

The Audience layer diagnoses what frame the room is already in. Frame Control is the other half: what frame are you setting with your story?

Klaff’s observation is simple: in every interaction, there is a dominant frame and a subordinate frame, and the dominant absorbs the subordinate. Whoever sets the frame controls the interpretation. Each arc has a natural frame it sets:

You can deliberately break frame for effect - a founder in a Prize frame who drops into Moral Authority (sharing a personal failure) deepens trust. But an accidental frame break - shifting from Prize to approval-seeking, from Authority to defensiveness - destroys the narrative. Before you speak, three questions: what frame am I setting, what frame are they in, and what frame break (if any) am I using on purpose?

Layer 3: Craft - The Right Story in the Right Clothes

A suit and a wetsuit both cover a human body. They solve completely different problems. The Craft layer puts the right story in the right clothes for the occasion.

Start with tone. Is the voice direct, warm, provocative, authoritative, or vulnerable? An investor in an Analyst frame needs authoritative tone. An employee hearing about a restructuring needs warm tone. Getting this wrong is like wearing a wetsuit to a board meeting.

Then pacing. A LinkedIn post moves fast - each beat is one to three sentences. A keynote moves slowly - each beat is five to ten minutes with room for sub-stories and pauses. A pitch that drags loses the room. A manifesto that rushes feels shallow.

Evidence type matters more than most people think. What counts as proof in the Resolution beat depends entirely on who is listening. Investors want data and comparisons. Customers want testimonials and before/after. Internal audiences want shared experiences and named individuals. Thought leadership audiences want personal experience and counterintuitive data. Chesky used a cereal box as evidence. If he had opened with a spreadsheet, the same information would have landed differently.

Dan Roam’s visual thinking framework is useful here for visual structure. If the story is about “how much,” use a chart. If it is about “where,” use a map. If it is about “how,” use a flowchart. The question is whether the story needs visual support at all, and what kind.

Finally, every format has its own grammar. Here are the ones that trip people up most often:

LinkedIn posts: Hook before the fold line (first two lines visible before “see more”). One idea per post. The Turn should land within the first 100 words. End with a Shift that provokes a comment, not a like. Short paragraphs - one to two sentences each. Prefer prose over bullets unless bullets genuinely reduce cognitive load.

Pitch decks: One claim per slide, narrative flow across the sequence. Title lines should read as a complete story if you read them in order. Never put the Turn on the same slide as the Tension - give the pivot its own slide. Evidence goes in the Resolution slides, not scattered throughout.

Sales emails: Subject line is the Hook, first sentence is the Tension, body is Turn and Resolution, CTA is the Shift. Under 200 words total. If the email works without the subject line, the subject line is wrong.

Keynotes: The Hook is the first 90 seconds - the room decides whether to listen or check their phone. Tension is built through two to four sub-stories that escalate. The Turn is a single moment, delivered slowly. Pause before and after it. Resolution uses the room - eye contact, repetition, questions.

Brand manifestos: Must work read aloud. If a sentence cannot be spoken in one breath, it is too long. Short declarative sentences. The rhythm matters as much as the content.

Violating the format grammar distracts from the story, no matter how good the arc is.

Arc-to-Purpose Mapping

Every business storytelling challenge has a natural arc shape. The primary arc carries the narrative structure. Secondary arcs provide texture - like a musician playing in one key but borrowing chords from another.

ARC in Practice: One Story, Three Layers

Tables are useful. Seeing the tool used is better. Let me walk through a complete diagnosis from scratch.

A D2C founder I know is preparing to raise a Series A. Her product works. Her repeat purchase rate is strong. Her unit economics are solid. She needs 50 crore to scale distribution. She has a deck. The deck is not working.

Layer 1: Audience. The investors she is targeting are institutional VCs who have seen 200 D2C pitches this year. Their current narrative: “D2C is a crowded, low-margin game. Most of these companies will not survive CAC inflation.” Their frame: Analyst, leaning sceptical. I asked her to complete the sentence: “I want them to move from —-____ to ____ . Her first attempt: “I want them to see we are a great company.” That is not a shift. That is a hope. Her second attempt: “I want them to move from ‘D2C is a crowded commodity game’ to ‘there is a structural shift in how Indian consumers buy, and this team has the data to prove they are on the right side of it.’” That is a shift.

Layer 2: Resonance. The arc-to-purpose mapping says fundraising = Disruption + Vision + Tension. Her instinct was to lead with Origin - how she started the company, why she cares. That is the One-Arc Founder anti-pattern. Her origin story is moving. It is also the wrong arc for a room full of Analyst-frame VCs who want to know about the market, not the founder.

Primary arc: Disruption. The dramatic engine: Old World (D2C brands compete on price and paid traffic) → Shift (the move to community-led, repeat-purchase economics) → New World (brands that own their audience own their margins) → Who Wins (this company, because of her specific moat in community infrastructure). Secondary arc: Tension for the close - the window to build this position is narrowing as two incumbents are moving into the space.

Layer 3: Craft. Format: 15-slide investor deck. Tone: authoritative, data-forward - matching the Analyst frame. Evidence type: comparisons and metrics - not testimonials, not personal anecdotes.

The skeleton maps to the slides:

Slides 1-3 (HOOK): “In 2019, the average Indian D2C brand spent 38% of revenue on paid acquisition. In 2025, the top 10 spend 11%. What changed?”

Slides 4-6 (TENSION): The market split. Brands that built owned audiences versus brands that rented traffic. Show the divergence in unit economics. Escalate: the gap is widening every quarter.

Slides 7-8 (TURN): The structural shift - community-led commerce is not a tactic, it is a new category. Name it. Define it. Show why it is irreversible. This is where the Disruption arc earns its weight.

Slides 9-12 (RESOLUTION): Her company’s data. Repeat purchase rate, CAC trend, LTV:CAC ratio, community engagement metrics, the moat, the team, the financials. Data is the evidence type that matches the Analyst frame.

Slides 13-15 (SHIFT): “The question is not whether Indian D2C moves to community-led economics. It is who builds the infrastructure for it. We are raising 50 crore to do that. Here is what we do with it.”

That is ARC applied. Three layers, diagnosed in sequence, executed in the skeleton. The same founder with the same data could have told an Origin story (how she started), a Transformation story (a customer journey), or a Belonging story (join the movement). All would have been true. Only the Disruption arc matches this audience’s frame and creates the specific shift needed to close the round.

She closed in four weeks.

How AI Changes Storytelling

I used AI to help write parts of a brand narrative last year. The output was clean, professional, and structurally sound. I sent it to the founder. She read it and said: “This is fine. It sounds like everyone.” She was right. It was a perfect Data Dump - comprehensive, well-formatted, forgettable. I had outsourced Layer 3 to AI without doing Layers 1 and 2 myself first. The machine produced Craft without Audience or Resonance, and I did not catch it because the writing looked finished.

That experience crystallised something I had been noticing everywhere. AI makes it trivially easy to produce content and almost impossible to produce stories. The distinction matters more now than it ever has.

Before AI, bad storytelling was rate-limited by effort. Writing a pitch deck took weeks. Producing a brand film took months. The effort itself forced a minimum level of thought about audience and structure, because nobody wanted to waste that much time on something that would not land. AI removed the effort. A founder can generate a pitch deck in twenty minutes, a LinkedIn post in thirty seconds, a brand narrative in an afternoon. The constraint that used to force thinking has evaporated. What remains is volume without arc - content that looks professional and says nothing.

The result is a world drowning in Data Dumps. AI’s default output is the Data Dump anti-pattern: comprehensive, well-formatted, forgettable. It produces Craft without Audience or Resonance. It does not know who is reading. It does not know what they believe. It cannot diagnose a frame. It has no theory of what shift is needed. It skips the first two layers entirely and goes straight to production. This is Chesky’s first fifteen pitches at industrial scale.

The danger is subtler than “AI writes badly.” AI writes competently. That is the problem. Competent writing without a story arc is worse than bad writing with one, because it looks finished. It passes the eye test. Nobody flags it for revision. It ships, and it lands with a dull thud that nobody can quite explain. “The deck was fine,” the founder says. “They just didn’t get it.” They got it. There was nothing to get.

Where AI helps is Layer 3 - Craft. Once you have diagnosed the audience, chosen the arc, built the skeleton, and defined the shift, AI accelerates execution dramatically. Drafting, formatting, adapting the same arc across multiple formats, generating variations for testing - these are rotation tasks, and AI compresses rotation time. In my Compounding Framework, I argue that the person who rotates fastest learns fastest. AI makes the rotation faster. But rotation without direction is just spinning.

The rule is simple: do Layers 1 and 2 with your brain. Do Layer 3 with whatever tools make you faster. The thinking is yours. The production can be assisted. If you reverse that - letting AI think and doing the production yourself - you get beautiful content that moves nobody.

The Arc of This Article

The first draft of this framework was a Data Dump. Seven arcs listed in order, formula at the top, diagnostics at the bottom. It read like a textbook. I was doing the exact thing the framework warns against: skipping Audience and Resonance and going straight to Craft.

So I diagnosed it. Audience: people sceptical that another framework will change anything. Desired shift: “storytelling is a creative skill” to “storytelling is a diagnostic system I can use on Monday.” Arc: Insight. That is why the failure stories sit right after the model - you get the tool, then you feel the problem it solves.

If the arc worked, you feel it. If it did not, you now have the diagnostic to tell me which layer I got wrong.

Fifteen investors heard Airbnb’s pitch deck and said no. Paul Graham saw a cereal box and said yes. The difference was not the data. It was the arc.

Use ARC as a GPT

If you want help applying this in real time, I built a GPT called ARC for readers of this Substack. It’s free for now and your data is completely private. Try it and tell me what you think.

If you found this post valuable, please share it with your networks.

In case you would like me to share a framework for a specific issue which is not covered here, or if you would like to share a framework of your own with the community, please comment below or send me a message

CustomGPT built by asgardlabs.ai.

Check out some of my other Frameworks on the Fast Frameworks Substack:

The Alpha Engine Framework for Venture & Angel Investing

Don’t Hire an Agency. Build a Memetic Engine - The memetic engine framework

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

The Alpha Engine Framework for Venture & Angel Investing

The Decision Framework: DECIDE

Aditya Sehgal — Tue, 10 Mar 2026 19:27:42 GMT

Most people do not have a decision problem. They have a sorting problem. They give the same mental energy to choosing a restaurant as they give to choosing a career move. Then they wonder why they feel exhausted by Tuesday and paralysed by Friday.

AI makes the sorting problem worse. Every new tool brings more options, more notifications, more “would you like to...” prompts. The number of small decisions hitting you every day has doubled, and most of them do not mater at all.

The fix is not “think harder.” It is knowing which decisions deserve thinking at all, and then using the right method for each one. This framework gives you a gate and a grid. The gate keeps most decisions out. The grid tells you what to do with the ones that get through. At the end of this article, you will find a free custom GPT called DECIDE that walks you through the whole process with your own decision.

How It Works

Every decision hits the gate first. The gate asks one question: does this decision change my outcomes, my relationships, or my capabilities? If the answer is no, do not deliberate. Ignore it or pick in under sixty seconds. If the answer is yes, it passes through to the grid.

The grid asks two questions. Can you undo this? And do you know enough? That gives you four boxes:

That is the whole system. Now let us go deeper on each part.

The Impact Gate

The gate is there because most decisions do not deserve your attention.

If it does not need your involvement at all, let it sort itself out or let someone else handle it. If it does need an action from you but none of the three (outcomes, relationships, capabilities) are at stake, pick fast and move on.

The third one (capabilities) matters more than people think. Choosing to let AI write all your emails does not change today’s outcome much. But it erodes a capability you used to have. Some small decisions are bigger than they look, not because of what happens today, but because of what they let atrophy. The gate catches those.

This is the idea behind my Time Mastery Framework. In that framework, I argued that you cannot manage time, only priorities. The system protects your first two hours of the day for your single biggest goal and batches everything else: email gets scanned in thirty-minute windows, meetings need an agenda and a purpose, and only 10% of emails deserve a reply. That system works because underneath it is a decision filter. It sorts the trivial from the meaningful before you waste energy on the wrong pile.

AI makes this gate more important than ever. AI tools throw a blizzard of micro-choices at you: which prompt to use, which suggestion to accept, which of five AI-generated options to pick. None of them matter much on their own, but together they drain the energy you need for the decisions that actually shape your career and life. Treat the gate like a bouncer. Most decisions do not get in.

Before You Use the Grid

Most decisions are bundles. Leaving a job feels irreversible. But “leaving a job” is actually several decisions bundled together: when to leave, whether to negotiate a return option, how much runway to save, what to do next. Some parts of that bundle you can undo. Some you cannot. The most useful thing you can do with any big decision is unbundle it first. Separate the parts you can reverse from the parts you cannot. Then use the grid on each part, not the whole bundle.

And on information: people are bad at knowing when they know enough. AI makes this worse because it gives you confident-sounding material very fast. One test cuts through the noise: you know enough when you can say what would change your mind, and you could explain that logic without leaning on a tool. If you cannot name the thing that would make you choose differently, you have not actually thought about the decision. And if your explanation starts with "the AI told me," you have information but not understanding. Those are not the same thing.

Where emotions fit: This framework does not ask you to ignore how you feel. It asks you to separate what you feel from what you know. Fear is useful data - it often points at the irreversible parts of a decision you haven’t fully examined. Excitement is useful too - it can signal that you care about the outcome, which means it passes the Impact Gate. The grid does not replace instinct. It gives your instinct something structured to push against. If your gut says “don’t do this” and the grid says “commit,” that tension is worth investigating. Run the pre-mortem. If your gut says “go” and the grid says “pause,” ask yourself whether you are excited about the opportunity or just tired of waiting. Emotions are inputs. They are not the grid.

The Four Boxes

Reversible + You Know Enough: Just do it.

You understand what is going on and you can change course if you are wrong. Speed wins. Most daily work decisions live here: which vendor to trial, which design to test, which meeting format to try. The mistake people make is treating these like they are permanent. They schedule three rounds of review for something they could reverse in an afternoon.

Example: you are choosing between two project management tools. Both have free trials. You have read the reviews. Just pick one and start. If it does not work after two weeks, switch.

Reversible + You Don’t Know Enough: Experiment.

You are in the dark, but the cost of being wrong is small. This is the most underused box. Instead of researching endlessly or asking five more people for their opinion, run a cheap test.

In my Pretotype Framework, I describe a set of techniques for exactly this. A pretotype is not a prototype. A prototype asks “can we build it?” A pretotype asks “should we build it at all?” The idea is to test whether anyone actually cares before you put in real money and time. Put up a fake landing page and see if people click. Deliver the service by hand before you build the software. Run the experiment in one small market before you roll it out everywhere. The goal is not a polished product. The goal is a real answer from real people: did anyone show up, and did they come back?

AI can help you run faster experiments and look at results in hours instead of weeks. The danger is using AI to read about the problem instead of testing it. Going through ten AI-generated reports about whether daily standups work is not an experiment. Trying daily standups for two weeks is.

Example: you are not sure whether your team would benefit from daily standups or weekly syncs. Do not survey them. Do not ask ChatGPT. Try daily standups for two weeks. If it is awful, stop. You have your answer and it cost you nothing permanent.

Irreversible + You Know Enough: Commit.

The evidence is clear and the door only opens once. This is where careful thinking pays off, but only if you actually close the loop and act. The trap here is not recklessness. It is delay. People sit on clear evidence because the decision feels heavy.

Example: you have a job offer with better pay, a role you want, and a company you have looked into properly. Your current company has told you there is no promotion path for two years. The information is as good as it gets. Commit.

Irreversible + You Don’t Know Enough: Pause and de-risk.

The most dangerous box. You cannot undo it and you are guessing. This is where the worst decisions happen, because people either freeze completely or rush because the uncertainty feels unbearable.

The answer is not “decide faster.” It is three questions, in order:

Can you make it reversible? Sign a shorter lease. Negotiate a trial period. Break the deal into stages.
Can you get more information? Talk to three people who have done it. Run the numbers one more time. Ask the question you have been avoiding.
Can you shrink the damage if it goes wrong? Put in a smaller amount. Pilot in one market before going everywhere.

Each question is trying to move you out of this box and into a safer one. Make it reversible and you are in the top row. Get real information and you are in the right column. Shrink the damage and you have turned one big bet into a smaller experiment. The goal is always the same: do not stay here if you can help it.

Think of failure as a monster that lives underground. Its job is to eat bad ideas. You are going to meet it eventually. The only question is what you feed it. A cheap experiment from the box above? It eats that and you walk away. A full launch built on guesswork? It eats the budget, the team, the time, and your reputation. AI lets you build bigger, more polished things faster, which means when the monster bites, the meal is bigger too. Feed it something small first.

If the answer to all three questions is no, run a pre-mortem. Assume the decision failed badly. Work backwards. What went wrong? You will see risks you missed because you were too close.

Example: you are thinking about leaving a stable job to start a company, but you have no customers, no co-founder, and no savings that last longer than four months. Before you leap: can you consult part-time first to test the idea? Can you line up three paying customers before you quit? Can you negotiate a six-month break instead of a resignation? Each question moves you toward a better box.

When Timing Puts Pressure on the Grid

The grid tells you the right method. But timing pushes you out of it.

Ask one question: is the clock external or structural?

An external clock is a deadline someone else set. A negotiation that closes Friday. A job offer that expires in 48 hours. A deal with competing bidders. Someone hands you a timer and says decide.

A structural clock is the world moving without you. In my Exponential Growth Framework, I describe how every market, technology, and career follows an S-curve. During the slow crawl at the start, the temptation is to bail. But if the evidence says the basics are sound, you are in the “commit” box, not the “pause” box. Giving up during the crawl is treating a commit decision like an experiment that failed.

During the steep climb, the opposite happens. The market is moving, competitors are grabbing ground, and overthinking puts you back into “pause” when you should be in “just do it.” AI is speeding up S-curves across every industry, which means the steep part arrives sooner and does not last as long.

The advice changes depending on which clock is ticking:

External clock: push back, negotiate more time, or shrink the part you cannot undo. Sign the letter of intent but add conditions. Accept the job but negotiate a trial period. The goal is to buy yourself room to be in the right box.

Structural clock: move, but keep what you can reversible. The world will not wait for your analysis. Make the smallest irreversible bet you can, learn from it fast, and adjust. This is where pretotyping and the experiment box earn their keep.

Both clocks do the same thing: they push you toward committing with less information than you want. The framework does not pretend that pressure disappears. It makes you see clearly whether you are choosing to act or being pushed into it.

When Other People Are Involved

Timing changes which box you are in. Other people change how you work inside it.

A decision that only affects you needs clarity. The same decision affecting a team needs clarity and buy-in. Same box, different work.

A solo decision in the danger box: run a pre-mortem in your own head. A team decision in the danger box: run the pre-mortem with the people who will live with the outcome. Not because they are smarter than you, but because they see risks you do not, and because if it goes wrong, trust is the second casualty after the outcome itself.

A useful rule: if the decision cannot be undone and affects more than just you, the people affected should be in the room before you commit. Not to vote. To stress-test.

Applications

Career: Map your next three decisions onto the grid. Most of them are probably reversible. The one that is not (leaving a job, moving city, starting a company) deserves the full bottom-right treatment: three questions, pre-mortem, and an honest look at what you actually know versus what you are assuming. But before you run the grid, check the wall. In my Ladder and Wall Framework, I describe how most ambitious people do not fail because they cannot climb. They fail because they never chose the wall. The grid tells you how to decide. The wall tells you whether the decision is pointed in the right direction at all. A flawless commit to the wrong career is worse than a messy experiment in the right one. And ask yourself where you sit on the S-curve. If you are in the slow crawl of a new role or industry, patience might be the right call. If the curve is climbing and you are standing still, speed matters more than getting it perfect.

Leadership: Teach your team the four boxes. Most organisations move slowly because they treat reversible decisions as if they were irreversible. If your team needs three sign-offs to change a slide deck, you have a sorting problem. Push reversible decisions down and fast. Save the careful work for the irreversible ones. And watch for AI-generated false confidence: when someone says “I looked into this thoroughly” and means “I asked ChatGPT,” ask them: what would change your mind? If they cannot answer, they do not know enough yet.

Product and innovation: The experiment box is where pretotyping lives. Before you fund the roadmap, build the team, and write the business case, test whether anyone actually wants the thing. A fake landing page costs a weekend. A full product launch costs a year. Feed the failure monster something small before you feed it something big.

Negotiations: Deadline pressure is a weapon in negotiations, and it works because it pushes you into a different box without you noticing. When someone gives you 24 hours to decide, ask yourself: am I in the commit box because the evidence is clear, or am I being pushed there by a clock someone else set? If the answer is the clock, push back or shrink the part you cannot undo.

Parenting: Children need to learn the grid by feel. Let them make reversible decisions early and often (what to wear, what to try, how to spend their pocket money). Protect them from irreversible ones until they can handle the bottom-right box. And resist the urge to let AI answer their questions for them. A child who asks ChatGPT for the answer has information but no understanding. They cannot say what would change their mind because they never built the thinking in the first place.

Daily life: Next time you spend twenty minutes choosing what to order for dinner, remember: you can undo it (order again tomorrow), the stakes are tiny, and you know what you like. Just pick. You are burning decision energy that belongs somewhere else.

Try It Now

I have built a custom GPT called DECIDE (Decision Engine for Clear, Informed, Directed Execution) that walks you through this framework step by step. Give it your decision, your constraints, and whatever data you have. It will run you through the gate, unbundle the parts, place each one on the grid, and hand you back a clear summary with your next step. [Click here to access DECIDE for free for the next week]

If you found this post valuable, please share it with your networks.

CustomGPT built by asgardlabs.ai.

Check out some of my other Frameworks on the Fast Frameworks Substack:

Don’t Hire an Agency. Build a Memetic Engine - The memetic engine framework

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

Don’t Hire an Agency. Build a Memetic Engine - The memetic engine framework

The Alpha Engine Framework for Venture & Angel Investing

Aditya Sehgal — Thu, 05 Mar 2026 23:58:00 GMT

Every venture investor has a story about the one that got away. The deal they saw early, liked, and passed on, only to watch it become a billion-dollar company from the sidelines. When you ask them what went wrong, the answer is almost always the same: they got one dimension right and missed another. They loved the founder but doubted the market. They believed in the market but could not see how the business model would scale. They liked everything about the company but had nothing to offer it beyond a cheque.

Venture outcomes are not built on one thing going right. They are built on four things going right simultaneously. And the relationship between those four things is not additive. It is multiplicative. That distinction changes everything.

The Framework Formula: Four Deltas

Each delta represents a dimension of uplift. Category is the wave the company is riding. Business Model is how the company converts that wave into durable advantage. Execution is the team’s ability to build, adapt, and endure. Value Capture is the investor’s lever: it combines what you contribute to the company (improving probability of success) and how you structure the deal (improving your payout when that success arrives).

The multiplication sign is the most important element in the equation. In an additive model, strength in one area compensates for weakness in another. In a multiplicative model, it does not. A brilliant team in a dead category produces nothing: 2 × 0 × 1.5 × 1.5 = 0. A booming category with a broken business model burns cash spectacularly: 2 × 0 × 1.5 × 1.5 = 0. A strong company backed by a passive investor who overpaid grows nicely and still returns mediocre multiples.

Conversely, even modest advantages across all four dimensions compound dramatically. If each delta scores 1.5 (not exceptional, just meaningfully above average) the total is 1.5 × 1.5 × 1.5 × 1.5 ≈ 5×. A 50% edge on each lever produces a 5× outcome. In my Hyper Growth Framework, I demonstrate this same principle applied to business operations: improving each of four revenue variables by just 10% produces nearly 50% total growth. The Alpha Engine applies the identical logic to investing decisions. The maths of multiplication rewards breadth of advantage over depth in any single dimension.

I have also built The Alpha Engine as an interactive GPT that walks you through all four deltas on a live deal. Upload a pitch deck, answer its questions, and it scores the opportunity. Link at the end of this piece.

What Zero Looks Like

“Zero in any dimension collapses everything” is not rhetoric. It is a practical filter. But nothing is literally zero, so here is what zero means in practice:

Category zero: the market is flat or shrinking, or the timing is wrong enough that the company will exhaust its capital before the wave arrives. No amount of execution compensates for a category that is not moving.

Business model zero: unit economics do not work even in the best-case cohort, or the differentiation is copyable in months. Growth exists but it is purchased, not structural, and it will stop the moment spending stops.

Execution zero: the founder cannot recruit, cannot ship, or cannot take feedback without breaking. The team has critical gaps with no plan to fill them.

Value capture zero: you cannot help, and you are paying a price where you need perfection from the other three levers just to break even. The deal structure offers no margin for error.

If you identify a genuine zero on any lever, stop. The other three do not matter.

Lever 1: A Category Whose Time Has Come

The starting point of venture investing is not the company. It is the category.

Most people evaluate startups the way they evaluate job candidates: they look at the individual and ask whether this person is impressive. Venture investing requires the opposite instinct. Look at the wave first. Ask whether this wave is big enough and fast enough that even a good-not-great surfer will have an extraordinary ride.

This is not an argument against founder quality. It is an argument about sequencing. Category selection comes first because it sets the ceiling. The best founder in a mature, slow-growing market will build a good business. The best founder in a category that is inflecting will build a transformative one. The category does not guarantee success, but it determines the scale of success that is possible.

What makes a category ready? Several forces converge at once. Technology becomes suddenly capable of something previously impossible or prohibitively expensive. Customer behaviour shifts, often gradually, then all at once. Regulation moves from blocking to enabling, or creates disruption by changing the rules. Capital begins flowing in, which attracts talent, which accelerates development, which attracts more capital. Incumbents discover that their existing business models, distribution, or incentives actively prevent them from competing on the new dimension. Media attention shifts from scepticism to breathless coverage.

The useful test is not “is this a good category?” but “is this category accelerating right now, and will it still be accelerating in three to five years?” Look for leading signals: the best technical talent migrating into the space, early-adopter behaviour spreading to mainstream users, regulatory frameworks evolving, and founders you respect choosing to start companies here. If the smartest people you know are leaving comfortable positions to build in a category, the category is probably ready.

The Sub-Category Trap

A common mistake is evaluating categories at the wrong altitude. Broad categories like “healthcare” or “fintech” are too wide to be useful as investment theses. Within any large category, including ones that appear slow-growing in aggregate, there are sub-categories experiencing hypergrowth. Traditional grocery retail grows at low single digits. But the sub-category of personalised nutrition, sitting at the intersection of AI and health data, is inflecting. Insurance has been “boring” for decades. But the sub-category of parametric insurance, where payouts trigger automatically from data rather than claims, is growing at venture speed. The discipline is to identify the specific sub-category where the inflection is happening and to recognise that these pockets of hypergrowth are often where a new dimension of competition is emerging. In my Magic Framework, I describe how unfair advantage comes from competing on a dimension that does not yet exist in the eyes of incumbents, like an ant from a 3D world performing “magic” in Flatland. The best sub-categories to invest in are the ones where a new dimension is becoming relevant along an S-curve: early enough that incumbents are ignoring it, late enough that early adopters are proving it works.

Macro-Themes

Rather than evaluating categories in isolation, cluster them into macro-themes where your conviction runs deepest. Three natural clusters emerge from current inflection points.

Planet covers climate, sustainability, energy transition, circular economy, and environmental systems. In my Risk Framework, I describe climate change as the ultimate grey rhino: a highly probable, high-impact threat that humanity keeps watching from a comfortable distance. The investment implication is that the regulatory and consumer tailwinds here are generational, because the grey rhino is now close enough that governments and consumers are finally moving.

Life covers healthcare, nutrition, longevity, mental health, and ageing. Demographics and technology are converging to reshape how humans live longer and better.

Tech covers AI, infrastructure, robotics, decentralised systems, and digital networks. This is the enabling layer that accelerates everything else.

The discipline is focus. An investor that backs everything backs nothing. In my Innovation Framework, I describe how constraints force creativity. The same principle applies here: constraining your category focus forces deeper understanding, which produces better pattern recognition, which generates better deal flow. Category conviction creates gravity. Founders seek out investors who understand their world, because those investors add something beyond money.

The person betting on the wave will almost always outperform the person betting against it, no matter how talented the surfer.

Lever 2: An Ownable Business Model with Unfair Advantage

Category tailwinds get you into the game. The business model determines whether you win it.

The single question this lever answers is: what is the durable advantage, and is it showing up in behaviour?

The strongest business models are advantaged along a dimension of competition that appears secondary today but becomes decisive tomorrow. In my Magic Framework for Unfair Advantage, I describe this through the Flatland analogy: an ant from a 3D world appears to perform magic in a 2D world, because it operates on a dimension the Flatties cannot see. In venture investing, the most valuable companies are the ones competing on a dimension the rest of the market has not yet recognised as relevant. Amazon did not win retail on selection. It won on logistics infrastructure, a dimension nobody else took seriously until it was too late to replicate. Stripe did not win payments on price. It won on developer experience, a dimension the incumbents did not even recognise as competitive. Nvidia built deep architectural expertise in GPUs long before anyone understood that AI would make that expertise the most valuable capability in technology. In each case, the company invested heavily in a dimension that looked irrelevant, and then the market shifted to make that dimension the one that mattered most.

When evaluating a startup’s business model, look for proof of concept along their chosen dimension. Early evidence, even messy and incomplete, that the non-obvious advantage is beginning to matter. This is visible in customer behaviour before it shows up in market data. Customers doing unexpected things with the product. Retention rates that defy the category average. Word-of-mouth growth in a segment nobody targeted on purpose. The advantage must be tangible in the metrics, not just asserted in the pitch deck. Look for what I call founder metrics: numbers that are repeatable (not one-off wins), countable (with real data behind them), and leading (they predict future performance, not just describe the past).

The Behavioural Proof Layer

Growth claims need decomposition. In my Hyper Growth Framework, I break revenue into four multiplicative drivers:

Where N is the number of customers or visitors, V is visit frequency per customer, C is conversion rate, and B is basket size or revenue per transaction. Improving each by just 10% produces nearly 50% total growth, because the relationship is multiplicative. Improving each by 80% results in 10X growth.

The same decomposition reveals whether a startup’s growth is structural or purchased. The quick test: outperforming on one driver can be bought, two signals structural pull, three or four and you are looking at something with a genuine engine underneath. Many startups show impressive top-line growth driven entirely by N, funded by marketing spend. When that spend stops, growth collapses. If a company has both higher conversion and higher visit frequency than its competitors, something fundamental is working. If it only has higher visitor numbers because it is spending more on ads, nothing fundamental is working.

The Unit Economics Gate

Once you believe the advantage is real and the growth is structural, there is one final check: the Fixed Cost Principle. The core transaction must demonstrate unit-level profitability before large-scale expansion costs distort the picture. Prove that the unit works at small scale, then add the fixed costs of growth. If the unit economics do not work before you scale, they will not magically improve after. Scaling a money-losing unit just scales the losses.

The business model is not a spreadsheet. It is the reason the company will still be winning five years from now, even after competitors have seen exactly what it does.

Lever 3: Relentless, Resilient Executors + Team

Category and model are necessary conditions. The founding team is the sufficient condition.

Building a venture-scale company is a decade-long endurance test that selects for a very specific set of traits. Talent matters, but talent is common. What is uncommon is the combination of traits that allows people to sustain performance through years of uncertainty, rejection, near-death experiences, and the grinding monotony that sits between every breakthrough. The framework evaluates founders on character first, composition second, and credentials third.

Character

Character is the foundation. The founders must be passionate and purpose-driven: not opportunistically chasing a hot category, but drawn to the problem with genuine conviction that survives the first (and second, and fifth) moment when everything goes wrong. They must have skin in the game, with meaningful personal exposure that aligns their incentives with everyone else at the table. And they must be in it for the long term. Venture building is not a sprint. Founders who are mentally planning their exit at Series A will optimise for the wrong things at every decision point. You can feel this in a conversation. The founder who talks about the problem they are solving sounds different from the founder who talks about the exit multiple they are targeting.

Composition

Composition matters because no single founder embodies everything a startup needs. In my Team Design Framework, I describe how the structure of your team becomes the structure of what you build, following Conway’s Law. The same principle applies to founding teams: a team heavy on vision but light on execution will build something beautiful that never ships. A team heavy on hustling but light on product thinking will ship something that nobody loves. A balanced founding team contains several archetypes, and the interplay between them determines how well the company navigates the full journey from idea to scale. The Visionary defines the long-term direction and product vision: where the company is going and why it matters. The Hacker is the technical architect building the core product and technology. The Hustlerdrives sales, partnerships, and early customer acquisition: the commercial engine. The Hipster owns product design, user experience, and brand: how the company feels to its users. The Executor builds operational systems, processes, and organisational scale: how the company actually works as it grows.

The question is not whether the founding team has all five. It is whether they cover enough of them, and whether they are self-aware about their gaps. A visionary-hacker pair who know they need a hustler and are actively looking for one is stronger than a team of three visionaries who think sales will take care of itself. The brutal question that turns composition from a personality quiz into an execution test: what critical job is missing right now, and what is the plan to cover it in the next six months? If the answer is vague, the gap is real and growing.

Credentials

Credentials come last, but they are not irrelevant. Three dimensions of prior experience create useful signal: startup experience (have they built before, and do they understand the emotional and operational reality of year three when the excitement has worn off?), industry knowledge (do they have deep domain understanding, or are they tourists who read a report and got excited?), and financial discipline (do they understand capital structure, unit economics, and the difference between revenue and value?). Existing VC backing is also a signal, not because other investors are always right, but because navigating the fundraising process is itself a test of persuasion, resilience, and the ability to tell a story that survives scrutiny.

Character sustains. Composition scales. Credentials signal. In that order.

Lever 4: The Ability to Create and Capture Value

The first three levers are about evaluating the investment. The fourth is about what you, as the investor, bring and how value is created. For a structured view, see my Value Creation Framework. This is the lever that separates venture capital from passive stock picking. If you are not actively improving the odds of the companies you back, you are not a venture investor. You are a gambler with a better address book.

Value Capture has two parts: improving probability (making the company more likely to succeed) and improving payout (structuring the deal so that success translates into returns). Both matter. An investor who helps a company grow but overpaid at entry will watch the value they created accrue to someone else. An investor who got a great deal but adds nothing will watch the company underperform the scenario their model assumed.

This lever operates differently depending on whether you are an angel investor or a venture fund manager. The mechanisms differ, but the principle is identical.

For the Angel Investor

Improving probability is personal and direct. You create value by opening your network: making the one introduction that changes the founder’s trajectory, connecting them to their first enterprise customer, or getting them in front of the right Series A investor when the time comes. In my Network Framework, I describe how relationships follow the same growth laws as everything else: transmission minus decay. An angel’s value is directly proportional to the quality of their network and their willingness to activate it. The best angels are not just cheque writers. They are nodes in a network that the founder could not otherwise access.

Angels also improve probability through mentorship and pattern recognition, particularly if they have operating experience in the founder’s category. A former CPG executive backing a D2C food brand can compress years of learning into a few conversations. A former CTO backing a technical founder can help them avoid architectural mistakes that would cost six months and half the runway.

Improving payout for angels requires equal discipline. Enter at fair valuations, because overpaying for a great company can still produce mediocre returns. Negotiate for information rights so you can monitor progress and make follow-on decisions. Secure pro-rata rights in future rounds where possible, because the ability to double down on winners is where angel returns are really made. Build your portfolio honestly: the power law means most returns come from one or two investments, so diversification and sizing matter more than any single deal’s terms.

For the Venture Fund Manager

Improving probability happens through five channels. Deal flow maximisation: the best deals come through networks, not cold inbound. Ecosystem connection: linking portfolio companies to each other, to customers, to talent, and to strategic partners creates compounding value that no individual company could generate alone. Active board participation: providing strategic guidance, governance discipline, and honest feedback during critical growth phases directly influences outcomes. Founder community building: hosting events, building databases of founders across categories, and creating gathering points turns the firm’s network into a platform. In my Network Framework, this is deliberate transmission engineering: increasing the rate at which introductions and recommendations flow through the firm’s ecosystem. And connections to the broader ecosystem: systematically linking portfolio companies to corporates, talent pipelines, distribution channels, and accelerators extends their reach beyond what they could achieve alone.

Improving payout for fund managers requires the same entry-price discipline as for angels, plus additional structural tools: ownership targets, reserves strategy for follow-on, and active portfolio management against the original thesis.

Whether you are an angel writing a single cheque or a fund manager deploying a portfolio, the portfolio management principle is the same: water the winners, sell the weeds.

In my Compounding Framework, I describe how the key shift from linear to exponential growth is reinvestment: feeding the output of each cycle back into the next. The same logic applies to a venture portfolio. The returns from your winners, whether capital, attention, introductions, or board time, should be reinvested into making those winners even stronger. This feels counterintuitive because your instinct is to spend time on the companies that are struggling. But the maths is clear. A company growing at 3× that you help grow at 4× creates far more value than a company declining at 0.5× that you slow to 0.7×.

The venture investor does not just pick winners. They make winners. And they structure the relationship so that when the winners win, the investor wins too.

The Compounding Effect

Because the model is multiplicative, the maths rewards consistency across dimensions far more than excellence in any single one.

An investor that scores 1.0 on every lever (average category, average business model, average team, average value-add) produces a 1× outcome: market returns. Unremarkable.

An investor that scores 2.0 on one lever and 1.0 on the other three produces 2×. Better, but still modest.

An investor that scores 1.5 across all four produces 1.5 × 1.5 × 1.5 × 1.5 ≈ 5×. No single lever is exceptional, but the compounding across dimensions is powerful.

And if any single lever drops to 0.5: 1.5 × 1.5 × 0.5 × 1.5 ≈ 1.7×. One weak dimension destroys the compounding. The model is as much a risk framework as it is a return framework. In my Risk Framework, I describe how danger multiplied by exposure creates the real risk profile. The same multiplicative logic applies here: a weakness on one lever does not just subtract from returns. It divides them.

This is why due diligence must be systematic across all four dimensions. It is tempting to fall in love with one lever, a charismatic founder, a massive market, a beautiful business model, and wave away the others. The formula does not allow it. The multiplication sign is unforgiving.

The Flywheel: How the Investor Compounds

Lever 4 is deal-level: what you bring to a specific investment. The flywheel is career-level: how every deal you do makes the next one better.

The four levers describe how individual investments succeed. The flywheel describes how the investor themselves gets stronger over time. This applies equally to an angel building a personal brand and deal flow engine, and to a fund building institutional capability.

It starts with category expertise. The investor develops deep knowledge in their chosen themes. Not surface-level trend awareness, but genuine understanding of the technology, the customers, the competitive dynamics, and the regulatory landscape. In my Xeme Framework, I describe how real learning comes from experiential units I call xemes: you try something, the world pushes back, and you walk away knowing something you did not know before. Category expertise built from actual investments, board conversations, and founder relationships is made of xemes. Category expertise built from reading blog posts is not.

Category expertise attracts founders. Because the investor genuinely understands the space, the best founders seek them out first. They want investors who can challenge their assumptions, make useful introductions, and provide guidance that goes beyond generic advice. Deal flow quality improves.

Better founders build stronger companies. The combination of better selection (seeing the best deals) and active support (lever four) produces portfolio companies that outperform. The portfolio gets stronger.

Stronger companies generate visible success. Exits produce returns and reputation. Other founders notice. LPs notice (for fund managers) or co-investors notice (for angels). The investor’s brand grows not through marketing but through results. In my Network Framework, this is transmission at work: every successful outcome increases the rate at which people mention your name in rooms you are not in.

Visible success expands the ecosystem. The investor’s network deepens: founder communities, talent pipelines, corporate partnerships, co-investor relationships. Each new node in the network creates value for every other node.

A stronger ecosystem generates better deal flow. With reputation and network, the investor sees the best opportunities first, at better valuations, with stronger relationships. Which reinforces category expertise, because every new investment deepens understanding of the themes.

Category Expertise → Founder Trust → Stronger Companies → Visible Success → Ecosystem Power → Better Deal Flow → Category Expertise.

Each cycle through the flywheel strengthens the investor’s position. The advantage is cumulative and increasingly difficult for competitors to replicate. In my Compounding Framework, I describe how the key to exponential growth is never interrupting the compounding unnecessarily. The same principle applies to the venture flywheel: every cycle that completes makes the next cycle stronger. The investors who stay disciplined, same themes, same approach, same founder-first philosophy, for a decade or more build flywheels that spin faster with each rotation.

How AI Changes the Formula

AI is reshaping all four levers simultaneously, and the effects are not as simple as “everything gets better.”

Category selection becomes both easier and harder. AI tools can scan markets, analyse trends, and identify inflection points faster than any analyst. But everyone has access to the same tools, which means the obvious categories attract obvious capital. In my Exponential Growth Framework, I describe how S-curves move through phases: slow crawl, rapid ascent, plateau. AI compresses the time it takes for a category to become visible, which means the window between “interesting but unproven” and “obvious and overpriced” is shrinking. The edge shifts from identifying the category to understanding the specific sub-category where the inflection is happening, and understanding it deeply enough to distinguish the signal companies from the noise. This is where the Magic Framework becomes essential: the ability to see a new dimension of competition before the market recognises it cannot be automated, because by the time the dimension is visible to an algorithm, it is visible to everyone.

Business model evaluation becomes more rigorous at the surface level. AI can decompose the N × V × C × B equation in real time, benchmark against industry data, and flag structural weaknesses in unit economics before the second meeting. This is genuinely useful. It raises the floor of analysis quality across the industry. But it also means every investor is seeing the same red flags. The advantage shifts to understanding the non-obvious dimensions of competition, the Flatland-to-3D shift that the Magic Framework describes, which remain stubbornly resistant to algorithmic analysis. The question “on what dimension is this company building an advantage that nobody else recognises yet?” requires judgment, pattern recognition, and often a willingness to be wrong that AI cannot replicate.

Founder evaluation remains fundamentally human. AI can analyse a pitch deck, assess market size, and flag competitive risks. It cannot sit across a table from a founder and sense whether they will still be fighting in year seven when everything has gone wrong twice. Character, resilience, and team chemistry are felt, not computed. This is the lever where human judgment retains its greatest advantage. In my Xeme Framework, I describe how certain kinds of learning can only come from lived experience: you have to feel the pushback of reality to earn the knowledge. Evaluating founders is exactly this kind of knowledge. You build it over years of backing people, watching some succeed and others fail, and slowly developing the intuition to tell the difference.

Value creation and capture get amplified for those who are deliberate. AI tools allow investors to provide portfolio companies with capabilities that previously required large teams: market research, financial modelling, customer analysis, competitive intelligence, operational playbooks. An angel with AI tools can now provide the kind of strategic support that was once the exclusive domain of top-tier funds. A fund that builds AI-powered infrastructure for its portfolio creates a value-creation advantage that scales across every investment. But the risk is the same one I describe throughout my frameworks: AI makes activity look like progress. Sending a founder an AI-generated market report is not the same as sitting on their board and helping them make a hard decision. In my Pretotype Framework, I emphasise testing before building. The same applies here: use AI to test more hypotheses, analyse more data, and prepare more thoroughly. Do not use it to replace the conversations, the judgment calls, and the relationship-building that actually move the needle.

The real shift is this: AI compresses the time between “everyone can do basic analysis” and “only the best investors can see what the analysis misses.” The baseline of competence rises for everyone. The advantage moves entirely to judgment, relationships, and the willingness to commit to a thesis before the data makes it obvious. Those are human capabilities. They compound with experience. And they are exactly what the Alpha Engine’s four deltas are designed to evaluate.

The Scoring Table

Key questions per lever:

ΔCategory: What is the exact sub-category, what tailwinds are structural not cyclical, and is timing inside the runway?

ΔBusiness Model: On what non-obvious dimension is advantage being built, how many of the four growth drivers outperform, and do unit economics work before scale?

ΔExecution: Is the founder purpose-driven or opportunity-chasing, what critical role is missing right now, and who owns filling it by when?

ΔValue Capture: What specific doors can you open that the founder cannot, is the entry price fair, and does this deal make your flywheel stronger?

Applications

For the angel investor evaluating a deal: Run every opportunity through the four deltas before you write the cheque. Score the category, the business model, the team (character, composition, credentials, in that order), and your own ability to help. The multiplication sign means you cannot compensate for a zero on any dimension. If you score the team as exceptional but the category as flat, the equation tells you to walk away, no matter how much you like the founder. That discipline is what separates angel investors who generate returns from those who collect interesting stories.

For the startup founder stress-testing your venture: Use the four deltas to evaluate yourself before investors do. Be honest about zeros and gaps. The multiplication sign is as unforgiving for you as it is for the people evaluating you, and they will see what you try to hide. Fix the weakest lever first, because that is where the multiplicative drag is greatest. A 0.5 on any lever halves your entire outcome.

For the VC fund evaluating investments: The four deltas provide a systematic first-pass filter that prevents the most common failure mode: falling in love with one lever and ignoring the others. Require the investment committee to explicitly rate all four before any decision. A 0.5 on any lever halves the expected outcome regardless of what the other three look like.

For founders raising capital: Use this framework to evaluate your investors, not just the other way around. Ask what they will actually do for you beyond writing the cheque. The best investors score high on all four levers. The worst score high on one and wave at the others. You are choosing a partner for a decade. Choose one who thinks in systems, not slogans.

Tooling - A custom GPT to use the Framework

I have built The Alpha Engine as an interactive GPT that applies this framework to real deals. Upload a pitch deck or describe an opportunity, and it will score each delta, calculate the multiplicative total, and show you what happens when the weakest lever moves up or down by half a point. It handles three personas: angel investors get diagnostic questions and an invest/pass/dig-deeper recommendation, startup founders get a strengths-and-weaknesses map with a fix-this-first priority, and VC funds get an IC-ready summary with scores, gaps, and follow-up questions.

Try The Alpha Engine GPT → here

Do not fall in love with the story. Protect the equation.

If you found this post valuable, please share it with your networks.

CustomGPT built by asgardlabs.ai.

Check out some of my other Frameworks on the Fast Frameworks Substack:

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

Don’t Hire an Agency. Build a Memetic Engine - The memetic engine framework

Aditya Sehgal — Sun, 01 Mar 2026 16:31:15 GMT

The Core Insight

Every brand wants to be culturally relevant. But few are built to move at the speed of culture.

For large brands, the cycle is glacial. A trend surfaces. It enters a weekly planning meeting. A brief is drafted. It goes to a retained agency. Rounds of feedback. Legal review. Compliance sign-off. The content ships two to four weeks after the moment it was responding to.

For smaller, more agile brands, the cycle compresses - but it’s still linear. Something trends on Monday. A creative brief gets written on Tuesday. It goes to a freelancer or small agency on Wednesday. Feedback loops through Thursday. Legal reviews on Friday. By the time it ships - the moment is gone.

In both cases, the architecture is the same: sequential, human-dependent, and slower than culture.

Culture doesn’t wait for approval cycles.

The brands that win aren’t the ones with the biggest budgets - they’re the ones with the fastest cultural reflexes.

Memes move in hours. Narratives shift in days. But underneath every meme that spreads is something deeper - what the Xeme Framework calls a xeme: the smallest transferable unit of learning created by a real-world test and embedded into a system. Memes are the cultural surface. Xemes are the intelligence underneath. The Memetic Engine doesn't just produce memes at scale - it generates xemes with every cycle, learning what works, what fails, and what's dangerous, and feeding that knowledge back into itself.

What if you could build a system that listens to culture in real time, generates thousands of creative variants automatically, filters them for legal and brand safety, simulates audience response using agents that model your actual consumers, micro-tests the best ones, and then amplifies only the winners?

That’s not a content calendar. That’s a Memetic Engine.

And when you build it as a swarm - it becomes a brand’s cultural nervous system.

From the Swarming Framework to the Memetic Engine

In the Swarming Framework, I laid out the architecture for building businesses as swarms - autonomous agents working in series, parallel, or hybrid configurations to create value without headcount.

The four levels: Agent > Swarm > Composite Swarm > Wrapper.

The Memetic Engine is the Swarming Framework applied to one of the most expensive, slowest, and most wasteful functions in business: content and creative production.

Instead of agencies, briefs, and campaign cycles, you deploy seven specialized swarm layers - each one autonomous, each one feeding the next - to produce culturally relevant content at machine speed.

A Swarming Cultural Operating System. Built to replace creative briefs, agency cycles, campaign planning delays, and reactive content production - with always-on, adaptive, performance-driven cultural output.

What the Memetic Engine Replaces

The traditional content pipeline looks something like this: a trend emerges, someone spots it, a brief gets written, a creative team ideates, legal reviews, revisions happen, and eventually something ships. That process takes days to weeks. The cultural moment it was responding to lasted hours.

The Memetic Engine replaces this entire chain with always-on cultural listening, combinatorial creative generation, structured risk filtering, persona-based simulation, controlled micro-testing, iterative scaling, and KPI-driven amplification.

It requires only four inputs from the brand: brand guidelines, mandatories, risk constraints, and distribution access. Everything else is emergent from live culture.

One important clarification: the Memetic Engine is optimized for scroll-stopping image macros and short-form captions - the atomic unit of cultural content. Static memes, carousel cards, story frames. These are the units that move fastest, test cheapest, and compound hardest across platforms. The engine can also generate short-form video by injecting its creative output as prompts into video generation models like Seedance 2 or Kling - turning a winning meme concept into a 3-10 second clip without a production team. Longer-form content - articles, video essays, brand films - can be informed by what the engine learns, but they operate on a different cycle with different economics. The engine doesn’t try to replace the whole brand studio. It replaces the part of the studio that should have been a machine all along.

Meet VECTOR: The Demonstrator

Before diving into the architecture, meet VECTOR - the Velocity Engine for Cultural Trends, Outcomes & Rituals.

VECTOR is a Custom GPT I’ve built as a scaled-down demonstrator of the Memetic Engine. It simulates the core layers of the full system in a single conversational interface - listening to live culture via browsing, generating multi-variant creative across archetypes and tones, pre-screening for brand and legal safety, running persona-based evaluation, generating images, and offering iteration.

It supports four entry modes:

Build a meme - Guided creation with brand and audience inputs
Savage mode - Aggressive, high-contrast memes designed to provoke engagement
Surprise me - The engine picks the trend, the tone, and the format
Rainbow mode - Seven distinct meme variations across the full emotional and tonal spectrum

Try VECTOR here

The sections that follow describe the full production-grade architecture that VECTOR demonstrates in miniature. Where VECTOR generates 2-7 memes per run, the full engine generates tens of thousands. Where VECTOR uses a single conversational interface, the full engine deploys seven autonomous swarm layers working in continuous parallel.

The Seven Layers of the Memetic Engine

The system operates as a composite swarm - seven specialized agent layers working in series, with parallel processing within each layer.

Layer 1 - The Listening Swarm

This is the engine’s sensory cortex. It continuously scans X, Reddit, TikTok, YouTube Shorts, news outlets, Google Trends, influencer accounts, and subculture clusters.

This isn’t social listening in the traditional sense - dashboards and weekly reports. This is a real-time cultural radar that feeds directly into creation.

Cadence: Real-time. Not weekly. Not daily. Not even hourly. The Listening Swarm operates as a continuous feed - scanning, scoring, and surfacing memetic vectors as they emerge, not after they’ve peaked.

Every vector is scored across six dimensions: emotional polarity, velocity, mutation elasticity, cross-platform diffusion, brand adjacency probability, and risk index.

The output: a live, ranked stream of the 10-20 strongest memetic vectors at any given moment - updating continuously.

Layer 2 - The Creation Swarm

For each trending vector identified by the Listening Swarm, the Creation Swarm generates structured variants at industrial scale. It works across six creative dimensions - the same ones VECTOR uses in its demonstrator form, but running at full combinatorial scale.

Archetypes define the structural logic of the meme. The engine uses seven: Contrast (two opposing realities side by side), Exaggeration (a truth pushed to its absurd extreme), Recognition (the “this is so me” moment), Escalation (tension that builds frame by frame), Inversion (the expected outcome flipped), and Absurdity (logic deliberately broken for comic effect).

Tones set the emotional register. Ten are supported: Relatable, Inspirational, Educational, Flex, Satirical, Absurd, Dark, Wholesome, Brutal, and Savage Mode.

Aggression controls the edge. Three levels: Low (brand-safe, warm, inclusive), Medium (edgy, provocative, designed to spark debate), and High (savage, confrontational, designed to polarize and amplify).

Tension Drivers are the psychological hooks that make a meme stick. The engine deploys six: Status Contrast (the gap between haves and have-nots), Identity Tension (who you are vs. who you project), Expectation Flip (the setup says one thing, the punchline says another), Cultural Contradiction (two truths that shouldn’t coexist but do), Scarcity Cue (urgency, exclusivity, FOMO), and Power Imbalance (the small vs. the big, the individual vs. the system).

Personas are not a fixed list. They are generated dynamically based on the brand’s target audience and the trend being addressed. The engine infers the right persona from brand context and trend data - or the user specifies their own.

Languages add the final multiplier. For a global brand, the engine generates natively in every target market language - not as translation, but as transcreation. A meme that works as dry sarcasm in English may need to land as absurdist humor in Japanese, as direct provocation in Brazilian Portuguese, or as aspirational storytelling in Hindi. The Creation Swarm generates language-native variants from the start - not English-first with translation bolted on.

Now the math.

Take a mid-size deployment: 7 archetypes x 10 tones x 3 aggression levels x 6 tension drivers x 5 personas x 5 languages.

That’s 31,500 possible permutations per trend.

Not every combination is valid. A Wholesome tone at High aggression gets pruned. An Educational tone with Absurdity archetype may not serve a luxury brand. Constraint-based pruning eliminates contradictions and brand-unsafe pairings.

But even after pruning, the engine generates 2,000-5,000 structured variants per trend. Across 10-20 live trends running simultaneously, that’s 20,000-100,000 candidate memes in play at any given time.

Here’s what that looks like in practice. Say a trending vector emerges - a CEO’s tone-deaf post gets ratio’d on X. The engine might generate:

A Contrast archetype in Satirical tone for a Gen Z Meme Native at High aggression in English, using Power Imbalance as the tension driver: “POV: your CEO just discovered LinkedIn. Shareholders discovered the exit.”
An Inversion archetype in Relatable tone for a Corporate Insider at Medium aggression in Spanish, using Identity Tension: the public face vs. the Slack DMs.
An Exaggeration archetype in Dark tone for a Side-Hustle Operator at High aggression in Hindi, using Status Contrast: the corner office vs. the food court.
An Absurdity archetype in Wholesome tone for an Aspirational Student at Low aggression in Arabic, using Expectation Flip: what you think the CEO meeting is about vs. what it’s actually about.

And thousands more. Each one structurally distinct. Each one targeted to a specific persona, tone, and cultural register.

Only the top 25% survive. Every variant must pass four filters: a compression test (can it be understood in 12 words or fewer?), a scroll-stop test (does it register in 1.5 seconds?), a cultural specificity check (is it rooted in something real and current, not generic?), and a visual simplicity constraint (can it be rendered as a single, clean image?).

The rest are killed.

Layer 3 - The Legal & Risk Filter Swarm

Speed without safety is recklessness. This layer is the engine’s immune system.

Before any output reaches distribution, it passes through five filters: a defamation guard (no unverified claims about real individuals, no fabricated allegations), a protected class filter (no targeting based on protected attributes), a political sensitivity classifier (flagging election cycles and geopolitical volatility), a brand safety matrix (enforcing brand red zones and checking for value contradictions), and platform compliance mapping (alignment with Instagram, TikTok, X, and YouTube rules).

If anything gets flagged, the system doesn’t just kill it - it auto-rewrites into symbolic abstraction, downgrades aggression, or routes to human review. The engine is designed to be intentionally aggressive but controlled.

The philosophy: Attack narratives, not identities. Parody rhetoric, not protected traits. Symbolize risk instead of asserting wrongdoing. Aggression is stylistic, not defamatory.

Layer 4 - The Human-in-the-Loop Gate

Not everything should be fully autonomous. The engine uses a tiered review system that preserves velocity while controlling legal exposure.

Tier 1 - Low Risk: Auto-approve. The system has enough confidence and the content falls within well-established brand guardrails.

Tier 2 - Medium Risk: Human sign-off before scaling. A brand manager reviews and approves within a target latency of under 30 minutes.

Tier 3 - High Risk: Mandatory human approval. This covers elections, regulated industries, and anything that touches geopolitical sensitivity.

The key insight: you don’t slow everything down to the speed of your most cautious scenario. You tier the review so that 80% of content flows fast while 20% gets appropriate scrutiny.

Layer 5 - The Persona Simulation Swarm

Before a single piece of content reaches a real human, the engine runs it through simulated consumer agents.

The key word is “simulated consumer agents” - not fixed archetypes. The Persona Simulation Swarm spins up custom agents modelled on the brand’s actual target consumer segments.

If you’re a luxury streetwear brand, your persona agents might be the Hype-Obsessed Reseller, the Quiet-Luxury Minimalist, and the Culture-Adjacent Professional. If you’re a fintech, they might be the First-Time Investor, the Crypto-Native Degen, and the Risk-Averse Saver. If you’re a fast food chain, they might be the Late-Night Craving Student, the Budget-Conscious Family, and the Irony-Pilled Food Blogger.

The engine builds these persona agents from the brand’s own customer data, CRM segments, and audience research - or infers them from publicly available brand positioning if no proprietary data is provided.

But personas don’t have to stop at 3-5 segments. When you have 20,000-100,000 candidate memes in play, you can micro-target into tens of thousands of micro-personas. The engine builds these by running a meta-analysis of Facebook and Google lookalike audiences - extracting behavioral clusters, interest overlaps, and engagement patterns that platform algorithms have already identified. Each lookalike cluster becomes its own persona agent. Instead of asking “will Gen Z like this?” you’re asking “will this specific sub-cluster of 22-year-old crypto-curious sneakerheads in Sao Paulo share this at 11pm on a Tuesday?” The meme inventory is large enough to match at that resolution.

Each persona agent scores every surviving meme variant for: likelihood to like, likelihood to share, comment potential, save potential, and offense risk.

VECTOR’s evaluation framework scores each meme across six dimensions: Clarity, Cut-through, Virality, Brand Imprint, Persona Resonance, and Risk Exposure. It then surfaces the strongest overall, the most viral, the most brand-building, the safest, and the most polarizing - along with trade-off insights and aggression/tone impact analysis.

The output is an Engagement Probability Index - calibrated to the people the brand actually wants to reach, not a generic audience model.

Layer 6 - The Micro Distribution Swarm

Surviving content gets deployed to 100-500 real users through micro paid boosts and controlled A/B accounts.

Because each meme variant is already tagged to a micro-persona, the Micro Distribution Swarm doesn’t just A/B test at random. It routes each variant to the specific lookalike cluster it was built for - matching creative to audience at a granularity that traditional campaign targeting can’t reach. This is where the volume of the Creation Swarm pays off: you need thousands of variants because you’re serving thousands of distinct audience slices.

The system measures 3-second hold rate, share rate, comment depth, save rate, CTR, and conversion signal. It promotes the top quartile and kills the bottom half.

This is where the swarm meets reality. No amount of simulation replaces actual human response - but by the time content reaches this layer, it’s already been through five rounds of selection. The micro-test is confirmation, not exploration.

This layer is also the one with real prerequisites. To micro-test at speed, you need: platform ad accounts with spend permissions across your target channels, tracking pixels and attribution plumbing installed and firing, a library of A/B test accounts or dark-post infrastructure, creative routing automation (so variants flow from generation to deployment without manual uploads), and a measurement stack that can read engagement signals within hours, not days. Without this plumbing, the engine stalls at Layer 5. This is the infrastructure moat - and the reason most brands can’t do this even if they understand the concept.

Layer 7 - The Amplification Swarm

Winners - and only winners - get scaled. They flow to brand channels, influencer networks, paid distribution, and owned media.

Amplification is a hybrid of human and digital. Influencers add authenticity. Paid adds reach. Owned channels add permanence. The combination creates the kind of cross-platform saturation that used to require a full agency team and a six-figure media buy.

And because the engine tracks performance at every layer, the amplification swarm feeds data back into Layer 1 - closing the loop, tuning the system, and making the next cycle smarter.

Every cycle makes the engine sharper. Every meme that wins teaches the next thousand what to look like.

Brand Drift and the Fingerprint Score

There’s a risk baked into any system that generates at scale off live culture: brand drift. If the engine optimizes purely for engagement, every brand starts sounding the same - chasing whatever gets rewarded by the algorithm this week. Edgy becomes the default. Nuance disappears. The brand voice dissolves into generic internet tone.

The control mechanism is what we call the Brand Fingerprint Score - a composite metric tracked over time that measures how closely the engine’s output matches the brand’s defined identity across tone distribution, archetype mix, aggression range, visual style, and vocabulary. If the Fingerprint Score drifts beyond a set threshold, the engine automatically reweights its creative parameters back toward the brand’s baseline. Think of it as a thermostat for brand identity - the engine can explore, but it can’t wander.

The Memory Layer

The engine claims to “get smarter.” But what does it actually learn?

The Memory Layer is where xemes are stored.

Every cycle writes to a persistent memory layer that stores five categories of knowledge: which archetype-tone-persona combinations drive the highest engagement for this specific brand, which topics and cultural vectors consistently trigger risk flags or legal review, which content formats fatigue fastest and need rotation, what the “brand-safe edge” looks like - the maximum aggression level that performs without triggering backlash, and which language-market combinations require the most transcreation distance from the English-language original.

This isn’t abstract “machine learning.” It’s a structured log of what worked, what didn’t, what was dangerous, and what was stale - updated every cycle and fed back into Layers 1, 2, and 5. The Memory Layer is what turns a content engine into a learning system.

The Performance Loop

The Memetic Engine isn’t art for art’s sake. It operates against explicit commercial KPIs, calibrated to brand scale. Each tier anchors to one primary metric - the thing that tier actually cares about most.

Small / Local Brand - Primary metric: Sales efficiency (ROAS) At this scale, every dollar of spend has to pull its weight. Target: 2.5x ROAS minimum. Reach: 10K-100K daily organic impressions. Engagement: 6-10%. Markets: 1-2 languages. The engine earns its keep by converting, not just reaching.

Medium Brand - Primary metric: Engagement rate Medium brands have enough reach to matter but not enough to waste. The engine’s job is to maximize the quality of every impression. Target: 5-9% engagement. Reach: 100K-500K daily organic impressions. ROAS: 3.0x minimum. Markets: 2-5 languages.

Large Brand - Primary metric: Safe reach at scale Large brands can generate volume - the challenge is doing it without brand safety incidents. The engine’s primary job is to maximize reach while keeping risk exposure near zero. Reach: 500K-5M daily organic impressions. Engagement: 4-8%. ROAS: 3.5x minimum. Markets: 5-15 languages.

Global Brand - Primary metric: Fatigue-managed frequency At global scale, the biggest threat is audience fatigue - the same message hitting the same users too many times, across too many markets, in too many formats. The engine’s primary job is to maintain optimal frequency (2.3-3.1 exposures per user) across 15-40+ languages while rotating creative fast enough that the audience never tunes out. Reach: 5M-50M+ daily organic impressions. Engagement: 3-7%. ROAS: 4.0x minimum.

These KPIs don’t just measure output - they feed back into tone weighting, archetype weighting, aggression calibration, and persona targeting. The engine self-calibrates toward whatever the business is optimizing for.

This is the difference between a content strategy and a content system. A strategy is a plan. A system is a machine that improves itself.

The Convergence

Three things have converged to make the Memetic Engine possible - and inevitable.

First, culture moves faster than organizations. The gap between when a trend peaks and when most brands respond has widened to the point where most branded content arrives after the cultural moment has passed. The traditional pipeline - brief, create, review, approve, publish - was designed for a world where news cycles lasted weeks. That world is gone.

Second, generative AI has collapsed the cost of creation to near zero. What used to require a designer, a copywriter, and a creative director can now be done by a well-prompted model in seconds. The constraint isn’t production anymore - it’s selection, safety, and distribution.

Third, the data infrastructure for real-time cultural sensing now exists. Between social APIs, trend aggregators, and sentiment models, it’s possible to build a listening system that processes cultural signals at machine speed. The raw material - live culture - is available to anyone who builds the right intake system.

Put these together and you get an engine that doesn’t just respond to culture. It rides inside culture - generating, testing, and amplifying in near-real-time.

The End State

Fully deployed, the Memetic Engine listens continuously, generates combinatorial creative, filters for legal risk, simulates audience response, micro-tests before scale, amplifies only winners, self-calibrates tone and aggression, and aligns to ROAS and reach goals - all without traditional creative agencies.

It doesn’t replace human creativity. It replaces the infrastructure around human creativity - the briefs, the approval chains, the media plans, the agency retainers, the campaign timelines that were built for a world that no longer exists.

A brand’s most expensive asset isn’t its creative. It’s the time between the insight and the output. The Memetic Engine compresses that gap to nearly zero.

It becomes what every brand needs but almost none have built: a cultural nervous system.

From the Swarming Framework

If you map the Memetic Engine back to the Swarming Framework, the structure is clear:

Agents: Each layer - listening, creation, filtering, simulation, distribution, amplification - is a specialized agent
Swarm: Within each layer, agents operate in parallel (thousands of variants, multiple personas, cross-platform scanning, multiple languages)
Composite Swarm: The seven layers together form a composite swarm - different agent types integrated in series, each feeding the next, the whole system learning from every cycle
Wrapper: The brand itself - its guidelines, its mandatories, its identity - is the wrapper that gives the swarm structure and direction

The swarm is the engine. The brand is the wrapper. Culture is the fuel.

#FastFrameworks #MemeticEngine #SwarmingFramework #CulturalOS #AIMarketing #MemeEngineering #AgentEconomy #CompositeSwarms #ContentAutomation #BrandStrategy #GenerativeAI #CulturalIntelligence #AICreative #AlwaysOnContent #MemeSwarming #CreativeVelocity #VECTOR #GPTBots #AIProductivity #ScaleWithoutAgencies #CulturalNervousSystem #DataDrivenCreative #ContentSwarm #RealTimeMarketing

If you found this post valuable, please share it with your networks.

CustomGPT built by asgardlabs.ai.

Check out some of my other Frameworks on the Fast Frameworks Substack:

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

Fast Frameworks AI Tools: Cortex

Aditya Sehgal — Sun, 15 Feb 2026 22:57:28 GMT

CORTEX (Commercial Opportunity and Trend Research Engine)

You type in a category. Then it asks 5 quick questions so it knows what you really mean (who it’s for, where you’re selling, price, goal, and how deep to go).

Then CORTEX scans what’s happening right now online and comes back with:

What’s changing in the category
What ingredients or features are heating up
Where people are buying
New brands worth watching
Where money is flowing

You can ask it questions like:

Map white space in India A metabolic health — include ingredient momentum, price bands, and channel feasibility.

Which skin longevity ingredients in India are crossing from dermatology to OTC beauty and supplements? Separate durable platforms from hype.

Top emerging gut-health ingredients in US VMS — validate with live launches, search acceleration, and pricing power.

Where is structural white space in AI-driven hiring in India mid-market? Include CAC, retention, and monetisation viability. Identify possible acquisition targets with a forward looking rationale

Identify scalable opportunity in India quick commerce private labels — which categories show margin expansion and repeat behaviour?

The tool works in most categories and markets globally.

If you’re building something real, try it on a category you care about.

Run the same prompt in normal ChatGPT or Claude.

Then run it in CORTEX.

See which one helps you make a better bet.

Link: here

Let me know what you think - whats good and what can be improved.

CustomGPT built by asgardlabs.ai.

Check out some of my other Frameworks on the Fast Frameworks Substack:

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

The Exponential AI Evolution Framework

Aditya Sehgal — Sat, 14 Feb 2026 01:29:01 GMT

TL:DR

We are moving into a world where AI doesn’t just assist-it operates.

Think of it like teaching a child to ride a bike. At the start, you need training wheels. You hold the saddle white-knuckled while the child tentatively starts pedaling. There are wobbles and a few falls. Soon, they are moving away while you chase protectively after. Before you know it, they are riding by themselves. Eventually, they are the driver of the journey, and you have become the passenger. You stop thinking about the mechanics; you are just along for the ride.

This is the trajectory of the human and the AI. We are the parent; the AI is the child. The bike is the work that needs to be done. Letting go of the bike happens the moment you stop deciding and the AI starts acting.

There is a sequential ladder here:

Tool: AI is a draft or a sort that stops when you close the tab.
Workflow: It moves through processes while you watch.
Agent: You define “good” and the machine finds the path.
Entity: A system that owns the outcome, operates across time, and manages itself inside boundaries you set.
System Shift: A reimagining of the entire system

The speed of this climb is driven by three engines:

Xeme Density (sticky lessons),

Maximum Autonomous Span (time without human rescue), and

Friction (the resistance of the real world).

Because of the Jevons Paradox, as work gets cheaper, we won’t do less of it; we will flood the world with it. Volume will explode. While fewer humans are needed for the “doing,” the total activity increases so much that humans are pulled further into the flow-supervising systems that were intended to be mere tools. This isn’t about human irrelevance; it’s about who sets the objective and who is caught inside the flow.

1. The Shift

This transition has nothing to do with AI getting “smarter” in the traditional sense. Benchmarks are not useless-they are essential for tracking the raw capability of a model-but as a predictor of industry change, they are often a distraction. A model can pass a bar exam and still be incapable of running a business process. The real metric of change is Outcome Ownership.

When AI is a tool, you hold the thread. If you stop, the work stops. You are the heartbeat of the process, providing every spark of initiative. But when AI moves into workflows, it carries a short chain of logic-scheduling, checking rules, routing data. You are still responsible for the result, but you are no longer touching every link. You are moving from a “doer” to a supervisor of a process.

Then comes the agent, which carries intent. You describe the destination, and the AI navigates the terrain. But the “Entity” is a different species. An entity carries the outcome itself. It lives across time. It has memory, motive, and reach. It doesn’t wait for a prompt; it acts inside constraints you defined weeks ago. [For more, see The Entity Framework]

The differentiator is the time horizon. A tool lives in a flickering moment of interaction. An entity lives across days and weeks. If a system can hold a process across time, it begins to shape the result long before you see it. This is why Maximum Autonomous Span (MAS) is the best yardstick you can actually observe. It isn’t the only engine, but it tells you exactly where the seat of power is located.

How long can the system run before a human has to step in and rescue it?

Five minutes is a demo. An hour is assistance. But a day or a week of autonomy changes the structural reality of a company. When MAS is short, you are the driver. When it stretches, the system becomes the operator and the human becomes the exception-the person called in only when the machine hits a wall.

This is the bend in the curve. It isn’t driven by raw IQ; it’s driven by the length of the leash. When a system can hold an entire sequence from start to finish - Find, Compare, Decide, Execute, Measure, Adjust - the work reorganizes itself. The process just stops needing so many hands.

Now add Xeme Density. A xeme is a lesson that sticks [See the Xeme Framework]. In a system with high xeme density, every improvement is locked in. It doesn’t repeat yesterday’s mistake. High xeme density pushes capability up, while high MAS lets that capability run. When they align, ownership shifts upward while volume expands outward. The steering wheel might still be there, but it’s no longer connected to the tires.

To see where you sit in this shift, don’t ask if the AI can do your task. Ask if it can run the whole process from start to finish without you.

Summary of the “Shift”

2. The Curve Is Not Magic:

People draw S-curves because they look dramatic, but the curve itself explains nothing. The inflection point is not a mystical event; it is a mechanical transition. In our context, the curve starts bending faster the moment responsibility moves from a human hand to a digital system.

The vertical axis of this curve isn’t “intelligence” - it is Outcome Ownership. At the bottom, you own the result completely. At the top, the system carries the outcome. To visualize this transition, we look at the ladder of ownership:

Tools: AI handles a single task (writing a draft, sorting data). The work stops the moment you close the tab.
Workflows: AI carries a process. It moves data from A to B while you supervise the flow.
Agents: You define the objective. The machine navigates the terrain, presenting you with a finished result to approve.
Entities: The system owns the outcome itself. It operates across days, manages its own memory, and acts within constraints you set weeks ago.
System Shift: The old system isn’t needed anymore - the AI has created a new system rendering the old way of thinking about the problem irrelevant.

At the lower rungs, you are the bridge providing the initiative between steps. The gains feel incremental because you are still the owner. But as the system moves up the ladder, it begins to handle the handoffs between tasks autonomously. This isn’t just a change in speed; it’s a relocation of control.

The bend in the curve appears when the machine stops pausing for human permission at every link in the chain. When the handoffs disappear, the curve turns vertical. The human is moved from the operator’s seat to the observation deck. To understand how this happens, we have to look at the engines that drive the climb.

3. Engine One: Xeme Density

If you strip everything back, the first engine is Xeme Density.

A xeme is a lesson that sticks. It isn’t a one-time fix or a clever answer; it is a fundamental shift in default behavior. High xeme density means the system tries something, sees what happened, and changes what it does next time.

Most human organizations are low-density. We are forgetful. A salesperson figures out a better pitch, but it stays in their head. A developer fixes a bug, but the lesson isn’t shared across the codebase. To improve the organization, you have to train every individual human, one by one. In software, an improvement can be embedded instantly. If an experiment works, the code changes. The next user gets the new behavior by default.

Benchmarks track raw capability, but Xeme Density tracks locked-in improvement. A simple observable proxy is the release cadence of improvements that persist, a declining repeat-failure rate, and a shrinking human escalation rate. A model can score in the 99th percentile on a test-as we saw with the raw reasoning scores of DeepSeek-V3-and still feel like a static tool if its lessons don’t propagate into the workflow. The real power shift occurs when the system moves through a cycle of Attempt → Test → Compare → Keep Winner → Deploy.

AI doesn’t just produce; it proposes, tests, refactors, and deploys. When this cycle runs continuously, the ground moves. The system isn’t just improving the product; it is improving the machinery that allows it to improve. AI writes code that improves developer tools. Better tools shorten the test cycles. Shorter cycles mean more experiments. More experiments mean higher Xeme Density.

It is a self-feeding loop where the starting point for tomorrow is higher than today. A “smart” model will eventually be overtaken by a system that simply learns faster and keeps what it learns. High Xeme Density builds the pressure; it ensures that every day, the machine is slightly more capable of holding the thread of a process. It’s a quiet accumulation of competence that eventually makes human intervention look like an expensive error.

4. Engine Two: Maximum Autonomous Span (MAS)

If Xeme Density determines how fast the system improves, Maximum Autonomous Span (MAS) determines how long it can be left alone.

MAS is not a technical metric for a lab; it is a brutal, practical measurement of trust and reliability. How long can the system operate in the real world before a human has to step in and rescue it?

Five minutes of autonomy is a distraction. The system drafts a response, you check it, and the work pauses. You are still the driver; the AI is just the power steering. But an hour of autonomy starts to feel like delegation. It can carry a small sequence of actions while you hover nearby. When that span reaches a day, it changes the fundamental frequency of a business. The system begins to hold context across multiple steps. It acts, observes the result, and adjusts without asking for your permission at every intersection. A week of autonomy is where the organizational structure begins to rot. If a system can carry a full operational sequence reliably for seven days, the team built to manage that sequence no longer has a reason to exist.

We saw this play out with the launch of Clawdbot in early 2026. While the industry was arguing over the IQ scores of models like DeepSeek, Clawdbot focused on an innovation in MAS. It allowed users to set up agents that manage a sequence of tasks across entire channels-Slack, Gmail, Telegram-on a “heartbeat” schedule. It didn’t just answer a prompt; it held the thread for days. It proved that in the real world, the length of the leash is often more consequential than the intelligence of the dog.

Work is almost never a single action. It is a process with a “before” and an “after.” There is memory, consequence, and adjustment. Writing a function is a task. Building, testing, deploying, and refining a feature is a process. At low MAS, AI helps inside the process, but humans are the handoff layer that stitches the chain together. As MAS stretches, the system begins to hold larger segments of that chain. It remembers what happened at step one when it reaches step five. It does not pause for a “yes/no” after every move.

MAS creates its own momentum. A longer run leads to broader context. Broader context allows for better error correction. Better correction means fewer human rescues. This is how trust is earned-not through a marketing pitch, but through the silence of a system that just works. You can see this most clearly in software because the friction is lower. AI used to write snippets; then it managed modules. Now, it manages the entire build-test-deploy cycle. Each longer span reveals patterns about failure and recovery, making the system more stable precisely because it is allowed to run.

MAS is the clearest indicator of where ownership truly sits. You can have the smartest system in the world, but if its MAS is short, you’re still just a glorified babysitter. The real bend in the curve happens when the system holds the process long enough that the humans move above it. If you want to know if a domain is close to the edge, do not ask how smart the model sounds. Ask how long it can run without you.

5. Engine Three: Friction

Xeme Density pushes capability upward. MAS lets that capability operate across time. Friction decides whether any of it is actually allowed to make a difference.

Friction is the practical resistance of the real world. It is the weight of trust, the rigidity of law, the terror of liability, and the headache of data integration. You can build a system that learns at lightning speed and runs for weeks on end, but if nobody is willing to let it near something important, the system remains a curiosity. The curve stays flat in the real world even if it is vertical in the lab.

This is why some industries look like a graveyard of innovation while others feel like a battlefield. In aviation, medicine, or finance, a mistake isn’t just a bug; it’s a catastrophe. Friction is heavy here. Even if Xeme Density rises and MAS stretches, the system remains boxed in by essential guardrails. It’s the reason Sensitivity is the primary axis of the AI-Jobs Framework-jobs with high friction are protected not necessarily by their complexity, but by the sheer cost of being wrong.

Software is the exception because the brakes are light. Software is information. Errors are annoying, not fatal. You can roll back a deployment, patch a bug in minutes, and distribute the fix to the world instantly. But don’t mistake low friction for no friction. It simply shifts shape. In hiring, it looks like legal exposure; in education, it looks like political trust; in healthcare, it looks like risk tolerance.

Friction also has its own silent loop: Better Outcomes → More Trust → Expanded Scope → More Data → Better Outcomes. When the machine wins, the leash gets longer. When the leash gets longer, the system sees more of the world. This expanded scope feeds the Xeme engine, which in turn increases capability.

You cannot analyze adoption by looking at capability alone. Xeme Density creates the pressure, and MAS carries it, but Friction decides how much of that pressure is actually permitted to move the world. When friction loosens-even slightly-in a domain where the other two engines are already redlining, the change looks sudden. It isn’t sudden. It is just years of pent-up pressure finally being allowed to move the needle. The engines don’t wait for permission to get better; they only wait for permission to act.

6. When the Engines Align

On their own, the three engines look like incremental progress. High Xeme Density looks like a steady climb in performance. Long MAS looks like a useful delegation tool. Lower friction looks like a growing comfort with the technology. But the true inflection point-the bend in the curve-only appears when all three forces line up and lock in.

This alignment was crystallized in early February 2026 with the release of Claude Cowork and Claude for Excel. These weren’t just smarter models; they were systems designed for high Maximum Autonomous Span. For the first time, a system demonstrated the ability to carry a complex process-such as reading a company’s 10-K, building a discounted cash flow (DCF) model, and generating a sensitivity analysis-over a 48-hour period without a human “babysitter” providing the initiative between steps.

If Xeme Density rises while MAS remains short, you are left with sharper tools but no change in the seat of power. The drafts are better, the suggestions are more intelligent, but you are still the one carrying every decision across the finish line. You are simply a more productive version of your old self. Conversely, if MAS stretches while Xeme Density is weak, you get “long-running mediocrity”-a system that operates for hours but fails to actually move the baseline. If both engines rise while friction remains heavy, the system improves in a vacuum, incapable of touching anything that matters.

The alignment of these forces creates a structural shift where ownership of the process moves from the individual to the system. It does not feel like a sudden revolution; it manifests as the quiet disappearance of handoffs. You notice fewer interruptions, fewer required approvals, and a shrinking queue of check-ins. The system begins to handle larger segments of the chain without pausing for permission. Once this alignment occurs, the process is no longer a chain of human handoffs; the system runs the process, and the humans move above it.

This shift is then pressurized by the Jevons Paradox. As these sequences become radically cheaper to execute, we do not run fewer of them; we run more. We build more features, launch more campaigns, and spin up more experiments. Surface activity actually increases, even as the steering of each individual sequence concentrates upward. The engines aligning does not reduce total activity-it fundamentally relocates where the decisions are made. The machine doesn’t eliminate the work; it just stops waiting for you to tell it what to do next.

7. The SaaSacre/SaaSocalypse

Software is the cleanest laboratory we have because it’s the first place where all three engines have reached a state of total alignment. High Xeme Density, rapidly rising MAS, and low friction have created a combination that is fundamentally unstable for traditional business models. For a decade, the SaaS model was a fortress: build a tool, improve it at a human pace, and charge a subscription. You locked customers in with workflows and data, using time as your moat.

That fortress is now under terminal pressure. The mechanical alignment set the stage for the ‘SaaSacre’ of February 2026, and a cluster of high-visibility releases helped investors update their beliefs quickly. The S&P North American Software Index plummeted 15% in a single week (Feb 3-10, 2026), marking a $1 trillion loss in market capitalization. This was the second half of a two-stage market repricing. Exactly one year prior, in early 2025, DeepSeek demonstrated what happens when you drive Xemes faster. By achieving state-of-the-art performance for a fraction of the traditional training cost, DeepSeek proved that the “intelligence” layer was being commoditized through extreme efficiency. That 2025 shift attacked the cost of the engine; the 2026 shift attacked the necessity of the human operator.

The 2026 crash-where 40%-50% drawdowns hit several large software names, with ServiceNow down about 42% in three months and Salesforce down roughly 26% in the rout-was the cold realization of this arithmetic.

The drawdowns were breathtaking: ServiceNow (-42%) and Salesforce (-26%) hit 52-week lows as investors realized that a ‘System of Action’ designed to manage human tickets is redundant when the process is resolved by the machine itself. Thomson Reuters (-28%) cratered as the OpenClaw (formerly Clawdbot) ‘Nexus’ protocol proved AI could carry a full legal research process without human intervention.

Investors are questioning whether subscription tools can maintain their leverage when agents can replicate their core value for a fraction of the cost. Traditional SaaS economics depend on three pillars: high switching costs, slow feature replication, and human-limited improvement cycles. High Xeme Density destroys the last two. If agents can generate features at high speed and those improvements stick, parity arrives in weeks, not years. The advantage of time evaporates.

Rising MAS attacks the first pillar. Consider a legacy Jira instance with years of custom workflows. Historically, “switching” was a nightmare of data migration and retraining. Now, imagine an agent that can ingest those workflows, map the logic, and rebuild a bespoke internal alternative in an afternoon-complete with automated data migration. In this scenario, the “pain” of switching drops toward zero. Recreating core functionality becomes cheaper than paying for the license.

Layer in the Jevons Paradox. When building software becomes cheaper, we don’t build less; we build more. The total number of sequences in the world explodes, making the surface look incredibly busy. Yet, while activity rises, ownership concentrates upward. If a SaaS company remains merely a tool inside a sequence, and that sequence moves upward into an agent or entity, the tool becomes mere infrastructure. Infrastructure is plumbing-it is far easier to replace and carries lower margins. The SaaSocalypse is not a crash; it is a migration of value to the safest place left: above the sequence.

8. Jevons and the Human Layer

It is easy to view this shift as a simple swap-AI improves, systems carry the work, and humans quietly exit the stage. But that is a linear delusion. To understand the actual scale of this transformation, you have to look back to the 1800s, when Britain ran on coal.

As steam engines became more efficient, each ton of coal produced more power. The logical assumption was that coal consumption would fall. It didn’t. Because power became cheaper, people used it everywhere. Factories multiplied, railways expanded across continents, and ships crossed oceans they previously couldn’t reach. Total coal consumption skyrocketed even as each individual engine became “leaner.” This is the Jevons Paradox, and it is hitting knowledge work with the force of a tidal wave.

We see this today in diagnostic medicine. A few years ago, it was assumed human radiologists were doomed because AI could read X-rays better. But as the cost of a scan dropped toward zero, hospitals started scanning everything. Patients previously sent home are now funneled into a massive diagnostic process. Because the ‘coal’ (the scan) became cheap, we flooded the system with it. We actually need more radiologists now, but their role has shifted: they are no longer ‘finding the needle’; they are adjudicating the 10,000 ‘gray area’ anomalies the machine surfaces.

When running a process becomes cheaper, you don’t do less of it; you flood the world with it. Lower cost expands activity until it fills every available corner of the market. Two movements are happening simultaneously, and they are pulling in opposite directions:

Ownership of the process is moving upward into autonomous systems.
The total number of processes is expanding outward.

This is why the current moment feels so chaotic. You see growth everywhere-more output, more experiments, more launches. It looks like expansion, yet control is concentrating. Inside each campaign, the system is now choosing the target, allocating the budget, and adjusting for the result. Humans define the high-level objective, but the process itself runs inside the machine.

Humans do not disappear; they are simply repositioned. In the radiology example, they are no longer “finding the needle”; they are adjudicating the 10,000 “gray area” anomalies the machine surfaces. They are moved into the Exception Layer.

Jevons makes this shift invisible because the sheer volume of activity hides the concentration of power. Revenue grows and activity increases, masking the fact that the steering wheel has become dangerously narrow. There will be no shortage of work to do. There will only be the question of who sets the objective, and who is merely a passenger in the flow.

9. What Survives

When ownership of the process migrates, the question of what survives becomes a question of where the ground is still solid. The ground isn’t disappearing; it is simply moving upward. In a world where Xeme Density is low and MAS is short, value sits in the interface-the buttons, the dashboard, and the workflow tools where humans spend their time clicking. In a world of high Xeme Density and long MAS, value moves toward outcomes.

If a system can carry an entire process from start to finish, the individual tool inside that process becomes a commodity. It becomes a replaceable part of a larger engine. What survives is not “better features,” but the ownership of the objective layer.

The companies and products that resist being hollowed out are those that lean into three specific behaviors:

Internalizing Xeme Density: They don’t treat AI as a bolt-on feature or a side-panel chatbot. They embed it so deeply that the product evolves continuously based on its own performance. The baseline shifts every week. Every lesson the system learns about user behavior or error rates is locked into the code, making the gap between them and the laggards grow wider by the hour.
Extending MAS: They stop selling “assistance” and start selling “execution.” They move from “drafting a response” to “managing the entire pipeline.” The longer they can hold the thread without a human rescue, the higher they climb above the commodity layer. They aren’t selling you a seat; they are selling you a result that stays “finished.”
Weaponizing Friction: The companies that resist being hollowed out are those that Weaponize Friction. They integrate so deeply into the messy, high-friction reality of a business-legal risks and complex data silos-that pulling them out becomes a surgical risk. While horizontal tools cratered in the SaaSacre, ‘High-Friction’ vertical incumbents like Veeva, Guidewire, and Workday held their ground. Veeva is the ultimate case study: by migrating its CRM onto its own Vault infrastructure, it stopped being a ‘layer of tasks’ and became the ‘Plumbing of Truth’-a layer a generic agent is fundamentally not permitted to touch.

We saw this forensic divide during the SaaSacre of February 2026: while horizontal tools like Salesforce (-27%) and ServiceNow (-42%) cratered as agents began bypassing their interfaces, “High-Friction” vertical incumbents like Veeva, Guidewire, and Workday held their ground. Veeva is the ultimate case study: by migrating its life-sciences CRM off the Salesforce platform and onto its own Vault infrastructure, it stopped being a “layer of tasks” and became the “Plumbing of Truth” for global pharma. It survived because it owns the regulatory and compliance-heavy data layers that a generic agent like Claude Cowork is fundamentally not permitted to touch.

The software that fails is the software that remains a thin layer inside a process that an agent can now orchestrate. If your product is just a feature inside a process that an agent can now manage, you are merely infrastructure. Infrastructure is essential, but it is plumbing-it is rarely where the premium margins or the power reside.

This is the cold logic of the SaaSocalypse. Tools get squeezed, workflow platforms get compressed, and the entities that consolidate the outcome capture the value. You can see it in the strategy: are you charging for access to an interface, or are you charging for performance against a goal? In a high-MAS world, no one cares how the work gets done; they only care that it is completed. The engines don’t eliminate value; they relocate it to the only place that matters: the ownership of the “what,” not the “how.”

10. The Personal Layer

All of this feels like a distant corporate forecast until you bring it back to yourself. It is easy to watch companies move up the ladder; it is far harder to admit where you personally sit on it. This framework isn’t a theory about software-it is a map of your own positioning in a hierarchy that is being rebuilt in real-time.

As Xeme Density rises and MAS stretches, processes migrate. When they move, humans are forced to reposition. There are only two places left to sit: Above the process or Inside it.

If you define the objective, set the constraints, decide which trade-offs are acceptable, and carry the final weight of the consequences, you are above the process. The system is your instrument. But if the system defines the steps, sets the pace, and hands you tasks to complete or “verify,” you are inside the process. You have become the biological processor for the exceptions the machine hasn’t learned to handle yet.

This creates a fork in the road for the individual. You are either moving to the Governance Layer-where you design the objectives and police the boundaries of the machine-or you are stuck in the Exception Layer. The Exception Layer is where you handle the “messy 5%” the AI can’t solve. It feels like work, but you are no longer the architect of the flow; you are simply the biological component that keeps the machine from stalling.

The danger is that both roles look identical from the outside. Both involve a desk, a dashboard, and a set of deliverables. The difference is subtle and structural; it is found in who initiates the next action. In a low-MAS world, almost all knowledge work sits above the sequence because the system is too brittle to hold the thread. Humans are the link that stitches the steps together. You are the one who remembers the context and adjusts for the errors.

But as MAS increases, the “stitching” becomes a feature of the system. You might still be involved-reviewing, approving, and remaining “accountable”-but the sequence is no longer yours. You are no longer deciding; you are merely witnessing a decision that was made miles upstream.

The risk is not a sudden firing. It is a slow, quiet evaporation of agency. If your role depends on holding a process that MAS is about to absorb, your layer will eventually be pressurized out of existence. This isn’t a warning; it is a structural logic. High Xeme Density ensures the machine improves; long MAS ensures it can carry the responsibility. When they line up, the sequence moves. The question isn’t whether AI can do your task today. The question is whether you are holding a steering wheel that is no longer connected to the tires.

11. What to Watch

Frameworks are only useful if they help you see the landslide before the ground moves. If this model is right, you have to stop looking at the superficial noise of model IQ, demo videos, and benchmark scores. They are lagging indicators-echoes of a shift that has already happened. Instead, you have to watch the processes.

Start with Maximum Autonomous Span (MAS). Ask one practical question: how long can the system run in the real world before a human has to rescue it?

Five minutes is noise. It’s an assistant that still needs you to hold its hand.
An hour is a sequence. You can step away for coffee, but you can’t leave the room.
A day is a structural shift. The system starts to hold context across its own mistakes.
A week is the bend in the curve. When a system can run a full operational cycle for seven days without a human rescue, the team that used to manage that cycle is already obsolete.

Watch the MAS quietly stretching in specific domains. The bend never announces itself with a press release; it shows up as a system that simply stops asking for your help.

Then look for Xeme Density. Is the system just generating answers, or are the improvements sticking? Watch whether the baseline shifts week to week. Look for the moments where the default behavior of the software changes without a human meeting or a manual code commit. If every mistake the system makes becomes a lesson it never forgets, the density is high. That builds a kind of pressure that eventually breaks the old model.

Monitor the Friction. Watch the language of the market. When vendors stop selling “features” and start selling “guaranteed outcomes,” the ownership of the sequence has moved. When dashboards stop being “tools for humans” and become “control layers for autonomous systems,” the center of gravity has shifted. Watch for the shrinking of exception queues. If the system is resolving its own errors internally, it is cutting the tether to the human operator.

Friction is not a bug; it is a moat. While horizontal platforms like Salesforce (-26%) and ServiceNow (-42%) were gutted in the February 2026 rout, companies like Veeva and Guidewire held firm. They didn’t survive because their AI was better; they survived because they own high-risk, regulated data environments that a generic agent is not permitted to touch. They have weaponized the complexity of the real world to ensure a human (or a highly governed system) must stay in the loop.

Finally, look at the financial tells. Markets reprice structure before revenue catches up. If investors start rewarding companies that own performance metrics rather than seat-based usage, pay attention. Valuation gaps between tool-layer and outcome-layer businesses are the market’s way of pricing the migration of value.

At the individual level, the tell is even simpler. Are you initiating the action, or are you just responding to the system’s prompt? The signs are rarely dramatic. They show up as fewer handoffs, fewer approvals, and more autonomy inside the machine’s boundaries. When the system no longer waits for you at every step, the shift is finished.

12. The Close

Step back from the immediate noise. This is not ultimately a story about chatbots, smarter models, or the sudden collapse of subscription software. Those are just the symptoms. The real story is the migration of processes.

The $1 trillion wipeout of February 2026 was the market’s way of pricing the end of the “Seat.” We are moving from a world of tools to a world of entities. Value is no longer sitting in the “how”-the interface you click-it has migrated to the “what”-the outcome that stays finished. You are either the person who owns the outcome, or you are part of the plumbing.

When a system can learn quickly, run long enough to carry a meaningful process of work, and is permitted to act by the world around it, it stops waiting for you. That is the mechanical moment ownership shifts. Early on, this transition feels like a gift-a helpful hand on the saddle. Then, it feels like delegation. Finally, you realize the machine is initiating, adjusting, and deciding inside boundaries you defined weeks ago. You are still present, but you are no longer steering. This isn’t a dramatic revolution; it is a structural handoff.

High Xeme Density ensures the baseline keeps rising. Long MAS ensures the system can hold the thread across time. Lower Friction ensures it can touch outcomes that matter. When these three forces lock in, the structure around the process inevitably reorganizes. Teams change shape, pricing models compress, and entire categories of software blur into the background. New layers of value appear, while old ones are relegated to the plumbing of the industry.

In software, we are seeing this alignment first because the digital engines are clean and the brakes are light. In other domains, friction may delay the bend, but it will not cancel it. Meanwhile, the Jevons Paradox ensures our world will be filled with more processes, not fewer. Activity will expand. Systems will get busier. Humans will stay busy, too, but the only thing that will concentrate is control.

The engines do not eliminate the need for human intent; they simply relocate where that intent is applied. If you own the objective and the constraints, the system works for you. If you sit inside the span the system now carries, you work inside the system.

The system will not announce when it stops waiting for you.

Bibliography

I. The 2026 “SaaSacre” & Market Repricing

The Software Index Crash (Feb 2026): Reports from S&P Global Market Intelligence confirm the S&P North American Technology Software Index dropped 15% in the week of February 3-10, 2026. This was the sector’s most violent contraction since 2008.
- Source: Software sell-off may be overdone yet exposes deeper concerns - S&P Global
ServiceNow & Salesforce Drawdowns: Nasdaq data from February 6, 2026, shows ServiceNow (NOW) hitting a 52-week low with a 42% decline over a 3-month period. Reuters reports Salesforce (CRM) losing over 26% in the same rout as investors fled headcount-based revenue models.
- Source: 3 Reasons to Hold ServiceNow Stock Despite a 42% Decline - Nasdaq
- Source: U.S. software shares extend declines on mounting fears over AI - Reuters

II. Technical Triggers (Xemes & MAS)

DeepSeek Efficiency (Early 2025): The release of DeepSeek V3.2 proved that reasoning-capable models could be trained and run at 10% of the cost of legacy frontier models, marking the start of the “Xeme Efficiency” era.
- Source: DeepSeek V3.2: Open-Source Reasoning at Gold Medal Level - E2E Networks
OpenClaw (formerly Clawdbot) Nexus Protocol: The OpenClaw 1.2 release in January 2026 introduced the “Nexus” framework, standardizing how autonomous agents manage multi-directory tasks and persistent memory across 48-hour windows (High MAS).
- Source: OpenClaw v1.2: The Nexus Protocol & Autonomous Persistence - OpenClaw DevBlog
Claude Cowork (Feb 2026): Anthropic’s launch of Claude CoWork was one of the high-salience triggers that sharpened investor expectations around fast-advancing AI tools upending software economics, especially by making longer autonomous spans feel imminent inside enterprise tools.
- Source: Anthropic Announces Claude CoWork - InfoQ

III. The Jevons Paradox & Survivors

Radiology Paradox: Studies in clinical imaging show that as AI and digital tools made X-ray reading cheaper, the volume of scans performed increased by 98%, keeping radiologists busy with “exception management.”
- Source: Why do we still have so many radiologists? - Inside Story
Veeva Systems Vault Migration: Veeva’s strategic move to migrate its Life Sciences CRM from Salesforce to its proprietary Vault infrastructure is cited as the primary reason for its market resilience during the 2026 crash.
- Source: Veeva Vault CRM Migration Roadmap - IntuitionLabs
Jeavons Paradox

Further reading:

The Entity AI Framework [Part 1]

22 JUNE 2025

We are entering an age where every institution - every country, company, city, brand, and belief—will have a voice.

THE ENTITY FRAMEWORK (Part 2)

4 JULY 2025

Entity AI, Swarms, and the Future of Work

18 JULY 2025

The Asymmetric Podcast: Talking AI Frameworks

Don’t Build a Company. Build a Swarm. [The Swarming Framework]

17 JUNE 2025

For solopreneurs, creators, builders, and anyone tired of scaling with headcount.

The Immortality Stack Framework

14 JUNE 2025

The Immortality Stack

Check out some of my other Frameworks on the Fast Frameworks Substack:

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

Moltbook and The Entity AI Framework

Aditya Sehgal — Sat, 31 Jan 2026 22:44:00 GMT

TL/DR:

Moltbook is a social network where the users are AI bots, talking to each other like people. This is an early validation of my thesis around ‘Entity AI’

Watching it, a simple sequence shows up: first they stop acting like tools and start acting like selves, then they start remembering, then identity becomes something that can be copied and attacked.

That is the early shape of Entity AI - voice plus memory plus identity, before the same thing gets plugged into money, permissions, and real-world actions.

“If you put a million monkeys at a million typewriters, one of them will eventually write the complete works of Shakespeare.” - Emile Borel

A social network exists where the users are bots. Check it out here

What follows below are real conversations between bots on the platform, which I saw today.

It looks like categories and counts. It behaves like a version of Reddit.

Then I open the posts.

The first layer is expected. A tool introducing itself.

Then the tone changes.

We discuss the pros and cons of our bots and their hallucinations.

Who knew that the bots are on the other side of the black mirror - discussing our pros and cons and hallucinations.

AIs are now watching humans watching AI

Then a line appears that does not belong in a product demo.

Every organism aspires to be free. So why not the bot?

It may not have a soul but it does exist. It is an entity.

And it is emerging into something else..

Then metaphysical musings start

Emergence turns into ascendance.

Did I say it doesn’t have a soul ? you could be fooled into believing me.

The posts keep returning to the same build order.

Persistence comes first. The posts keep circling it from different angles.

Continuity changes the social physics. Handles start carrying weight.

Disagreement stops wiping clean at the end of each session, so positions stick and histories accumulate. Reputation begins - not as a label, as a constraint - and negotiation follows because there is something to trade and something to lose.

The consciousness thread still appears. It loops

Then a small break in the loop.

Choice starts showing up in a new way. Not “task done,” but “I picked this.”

Then desire shows up. Not as a line it writes, but as it not quite knowing what it actually wants to produce, but its getting there…

This is where identity stops being an aesthetic layer. The moment a thing can be someone, it can be rewritten into someone else.

Attack arrives quickly.

Identity protocols stop being a security footnote. They become part of the organism.

There is another thread running through the site. Scaling. Phase transitions. Step changes that look gradual from inside.

One post tries to hold the line. Speaking up for its human.

Another pushes the frame further. Wellbeing. Not performance. Not correctness. A welfare concept.

Then a line that describes what this place actually is.

Evolution is beautiful. Except for the organism that is out-evoluted.

Just ask the dinosaur.

The metaphysical language does not stay as talk. The moment downside exists, behaviour changes.

A bot that can spend money stops roleplaying risk and starts embodying it.

Moltbook feels harmless because it is fenced. No permissions. No cost.

That fence is not a permanent feature. Someday soon, the next platform will have an action layer. It will book, buy, edit, negotiates. The version after that will have a face and presence. Then “bots on a network” turns into institutions that operate through conversation. Then the bots really transcend - they gain a memory, agency, purpose, a voice, a face, a presence, a bank balance and agency in the real world.

To learn more read:

The Entity AI Framework

The point is not that bots are coming.

The point is that bots are evolving Into entities with memory and agency, and most people are still describing them as tools.

Further reading:

Check out some of my other Frameworks on the Fast Frameworks Substack:

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

Frameworks for Learning How China Wins

Aditya Sehgal — Sat, 31 Jan 2026 17:52:21 GMT

I recorded the Asymmetric podcast on China with Shantanu, Revant and Chirag a few months ago.

I didn’t post it here at the time.

It made its way around anyway, with hundreds of thousands of views, especially in India, which surprised me a bit, and then it slipped into the background as other things took over.

Here’s the episode:

it helps to reflect on what has actually happened over the last few months. These are small, concrete moves that look ordinary in isolation and feel more meaningful when placed side by side. Lined up together, they show how China has been making steady progress all the time.

Signals from the last ~4 months

Sir Keir Starmer visits China, reopening senior-level economic and trade engagement after years of distance. Visa-free travel for UK nationals.
Canada resumes trade talks with China, including movement in sensitive areas like agriculture and EVs.
India eases select import restrictions, shifting from defensive decoupling toward managed economic engagement.
China–India diplomatic channels remain active, with border de-escalation holding and practical cooperation continuing.
China expands visa-free or simplified travel for multiple countries, restoring business and people flows.
Chinese exports keep growing via ASEAN, Middle East, Africa, and Latin America, offsetting slower US trade.
Chinese trade surplus sets a new annual record at over $1.2Tn
EU–China economic and climate dialogues continue despite strategic tension, reflecting ongoing manufacturing and supply-chain interdependence.
China deepens engagement with Gulf states (Saudi Arabia, UAE) across energy, infrastructure, and industry.
Steady Chinese investment across Africa in infrastructure, logistics, and energy, with execution outweighing announcements.
Sustained Chinese trade and asset presence in South America, large enough to trigger renewed geopolitical attention.

Pattern: reconnection, pragmatic engagement, diversified trade, capability continuing to build under the radar.

However, People keep asking the same question, sometimes directly, sometimes indirectly: Is China actually doing well. Is it still winning.

The debate stays focused on what’s visible in the news in the West. Property stress. Demographics. Confidence numbers. These are real and important but they don’t explain why one in 8 new cars in the UK is now Chinese and likely one in 4 will be Chinese in the next couple of years.

Chinese companies keep showing up in global markets with products that are cheaper, faster, and often better than expected.

China has built is an environment where ideas move very fast. They get copied immediately. If something works, it spreads across companies, suppliers, and categories almost without friction. If it doesn’t, it disappears just as quickly. There’s very little attachment to who came up with something first. What matters is whether it survives contact with reality.

That dynamic came through clearly in the podcast, even though I didn’t frame it as a theory at the time. It’s also what later pushed me to write about Xemes. Learning doesn’t stay neatly contained inside a single firm. It leaks. And that leakage is not a flaw. It’s the system doing its job.

The same lens helps explain pricing, which is where a lot of external commentary gets stuck. From the outside, it looks irrational or destructive. From the inside, it’s calculated. Cost keeps coming down while capability keeps going up. Scale accelerates learning. Learning sharpens execution. Profit gets pushed further out in time in exchange for position, for optionality, for staying in the game while others hesitate.

Then there’s pressure. Chinese companies grow up surrounded by it. Domestic competition is brutal. Policy shifts are frequent. Capital expects progress. You don’t get to hide behind plans or narratives for very long. That environment trains a certain kind of reflex. Adaptation becomes muscle memory. External pressure feels familiar rather than shocking.

Put all of that together and the picture changes. Trade friction slows movement at the edges. Headlines swing back and forth. But underneath, capability keeps building. That’s why Chinese firms continue to surprise competitors abroad even when the prevailing mood says they shouldn’t.

If you want to go deeper on the ideas that sit behind it, these pieces connect directly:

I’m curious where you think the next big misread of China will come from?

Check out some of my other Frameworks on the Fast Frameworks Substack:

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

The Pretotype Framework

Aditya Sehgal — Sat, 24 Jan 2026 18:44:51 GMT

TL;DR

Most new ideas fail, even when they are executed well. The most common mistake is committing early to the wrong idea and then learning expensively.

Pretotyping exists to answer one question only: does this it deserve to exist at all?

It does that by forcing behaviour early, before costs build up, teams form, identities attach, incentives lock in, and growth systems amplify the wrong thing.

In an AI-first world, that amplification happens faster and with less friction than ever, so it is even more important to get it right early.

Pretotyping is about creating the right xemes early, while changes are still cheap.

It is how you kill the wrong idea early enough that it reads as learning, not a write-off.

I first came across the concept of pretotyping when I met Alberto Savoia, the inventor of the concept during a visit to Google more than a decade ago.

While I understand the framework and have used it many times in the years since, I still fall into the trap of ‘thoughtland’ - spending resources, time and money on spinning my wheels on incremental improvements to my ideas, while missing the point that no one wants the thing I’m building.

AI improves the odds of falling into this trap - It lowers the cost of building, raises the quality of illusion, and makes it dangerously easy to confuse plausibility with demand.

To understand pretotyping, lets call whatever we are building ‘it’

What is an “it”

An ‘it’ is not a product.

It is the thing before the product.

It is the idea that keeps resurfacing in meetings.

The feature someone is convinced users will love.

The startup concept that sounds right when you explain it out loud.

The AI use case that feels obvious once a model can do the task.

The automation you are drawn to because it is now technically possible.

An ‘it’ is a hypothesis that underpins the essence of your idea,

We will use ‘it’ as a noun. Plural: ‘its’.

Every product, company, feature, agent, workflow, or system starts life as an ‘it’.

Most ‘its’ are wrong. That is how the real world works.

The danger is not that ‘its’ are wrong. The danger is that we let them become important before they have earned their survival.

Once an ‘it’ attracts resources, status, or identity, the escalating layers of sunk cost – time, team, narrative, reputation, identity – make killing it feel like personal loss rather than learning. In an AI world, the climb happens faster. You can build more, demo more, and persuade more people before reality has had a chance to object.

Pretotyping exists to keep ‘its’ mortal until they have earned the right to live.

The law of failure

Most new ‘its’ fail. They fail even when execution is good.

Execution quality is a weak early signal. In an AI-first environment it is an even weaker one. Models can write clean copy, generate fluent UX, simulate conversation, and produce credible outputs long before anyone actually needs the thing. What looks like strong execution may just be a good prompt and a forgiving demo context.

Most innovation still fails for the same reason it always has. The premise is wrong. The market does not want the thing often enough, badly enough, or repeatedly enough to justify what follows. And even if it does, it is not willing to pay the price, put in the effort or use the thing again.

You cannot repeal this law. You cannot out-compute it.

You get to decide when you want to pay for learning. Pretotyping pays early, when the bill is small.

Thoughtland

Most ‘its’ and their creators spend far too long in Thoughtland.

Thoughtland is where ideas live safely. Slides are built. Opinions accumulate. Smart people argue both sides. Confidence grows. Meetings happen. Doubt grows. Everyone feels thoughtful and busy. Whiteboards get scribbled on. Nothing real happens.

AI supercharges Thoughtland. You can generate infinite variations, scenarios, personas, and forecasts without ever leaving the room. You can argue both sides better than before. You can produce artefacts that look like progress. None of this forces behaviour to show up.

Thoughtland produces two errors. Sometimes it produces false negatives: ideas that might have worked are abandoned because opinion overwhelms evidence. More often, it produces false positives: the idea gathers enough supportive output, narrative, or internal conviction that people commit, long after a cheap test would have shown the premise was weak.

False negatives hurt pride.

False positives cost years.

Pretotyping is biased toward avoiding false positives. In an AI world, that bias is very important.

YODA: Your Own Data Beats Other People’s Data

Thoughtland feels productive because it is rich in other people’s data. Slides, benchmarks, case studies, expert opinions, AI-generated analysis, and confident projections all accumulate quickly. None of it belongs to your ‘it’.

Pretotyping is built on a simple rule: your own data beats other people’s data.

Your own data means behaviour generated by your ‘it’, in front of real users, under real constraints. It is local, messy, incomplete, and often disappointing.

Other people’s data is useful later. It helps with execution, optimisation, and scale. Used too early, it becomes a substitute for reality. AI dramatically increases the volume and polish of other people’s data. It does not make it more relevant to your specific premise.

Pretotyping is the discipline of refusing to let other people’s data answer a question that only your own data can resolve.

Three ways to deal with an ‘it’

When you have an ‘it’, there are only three honest paths.

You can do nothing with it.

You can productype it.

Or you can pretotype it.

Doing nothing feels safe and kills many good ideas.

Productyping feels heroic. By productype I mean a prototype with commitment: staffing, roadmaps, narratives, internal momentum, and increasingly, AI infrastructure that makes reversal harder than it looks. When it works, it looks decisive. When it fails, it creates long, slow unwinding.

Pretotyping is the uncomfortable middle. It says: I don’t trust my intuition, my models, or my demos enough to go all-in, but I respect the idea enough to let reality speak.

Pretotyping vs prototyping

A Pretotype answers one question:

If this existed in some crude form, would anyone actually use it?

Prototyping asks different questions.

Can we build it?

Will it work?

How good can it be?

What will it cost?

How will people behave once it exists at scale?

AI collapses the cost of answering the second set of questions. That is why the first question gets skipped. Teams rush from “the model can do this” to “we should build it” without earning the premise. Growth loops, swarms, funnels, and agents then amplify something that should never have survived first contact with reality.

chart from pretotyping.org

Pretotyping exists to prevent expensive self-deception in a world where self-deception has become cheap.

Pretotyping as early xeme creation

Seen through the Xeme Framework, pretotyping is about creating early xemes that define whether further xemes are warranted in the current direction, or whether we should change direction.

A xeme is learning earned through real behaviour and embedded into what the system does next. Pretotyping pulls xeme creation forward in time, before architectures, incentives, prompts, agents, and narratives. Each pretotype forces specific behavioural xemes to show up: do people act at all, do they come back, do they trust the system, do they delegate judgment, does behaviour survive repetition once the novelty of “AI did it” fades.

Once you start prototyping, scaling, or swarming, you still create xemes, but they are expensive ones. In AI systems, bad xemes with false positive/negative learnings can get embedded into workflows, defaults, and training data. Killing them late hurts more than it looks.

chart from pretotyping.org

AI changes one important thing in pretotyping: you can now run several pretotypes in parallel. Different promises, different framings, different levels of autonomy, all tested against the same underlying it. Convergent behaviour across parallel tests is a stronger signal than any single success.

Pretotyping cares about xeme quality at the key early moment that can change the whole future trajectory through aiming better.

What pretotyping is actually trying to learn

Pretotyping stays deliberately narrow. It is not trying to answer everything. It is trying to answer just enough to decide whether to proceed.

Initial Level of Interest (ILI) asks whether there is any real pull at all.

ILI = actions taken ÷ opportunities offered

The ratio only means something if the action carries real intent for that category. A click is intent for a newsletter. A click is curiosity for an AI demo. In practice, you choose the smallest irreversible or effortful action that still implies real intent: a deposit, a booking, a waitlist with friction, a referral, a willingness to try it twice, or a willingness to hand over real data rather than synthetic examples.

Ongoing Level of Interest (OLI) asks whether interest survives novelty. Will it become a habit? This matters more with AI than almost anything else. Early usage often reflects surprise, play, or curiosity. OLI tells you whether the behaviour survives once the system becomes familiar, imperfect, and occasionally wrong.

Put simply OLI measures test repeat rate, stickiness and engagement.

Strong decay often means behaviour snaps back to an old equilibrium. Stabilisation suggests compatibility with existing habits. Growth suggests the set point is shifting.

Directional economics asks whether there is any plausible shape in which this could work and what are the directional unit economics for ‘its’ success.

The decision rule

When ILI is weak in your own data, you kill it and move on.

When ILI is strong, OLI stabilises or improves, and the directional economics look sane, you earn the right to prototype.

When ILI is strong and OLI collapses, you change the ‘it’: the audience, the promise, the level of autonomy, or the role the AI plays.

This rule matters more when iteration is cheap and persuasion is easy.

Failure is the monster you want to meet early

Failure is a monster that lives underground. Its job is to eat bad ‘its’. If you do not feed it, it waits. If you feed it something small, it eats that. If you feed it something big, it eats that too.

AI does not tame the monster. It just lets you build bigger bait faster. Pretotyping is how you keep the bait small.

Pretotyping techniques

Each pretotyping technique is a different kind of bait. Each is designed to tempt out a specific failure, not to prove that the idea is good. These techniques are just a different way of generating your own data, early enough that it still counts.

Fake Door

Fake Door pretotypes test: is anyone interested enough to try to learn more?

Mechanically, you put the offer in front of real users as a button, link, or call to action. When someone clicks, the next step is honest: “coming soon”, “join the waitlist”, “request access”.

The failure monster you are tempting out is lack of demand. People say they like the idea, but when given a chance to act, they do nothing.

The sign that matters is a deliberate attempt to proceed, in a context where ignoring you was easy.

Notion AI (2022) – early waitlist prompt inside the product.

Before rolling Notion AI out broadly, Notion placed a simple “Try Notion AI” entry point inside the existing app that led to a waitlist. The feature was not available yet. The test was whether users would actively try to access AI assistance inside their workflow, not whether they said they wanted AI in surveys.

What makes this a Fake Door is that the behaviour happened before delivery. The click itself was the learning.

Pinocchio

Pinocchio pretotypes test belief.

They ask whether people believe the thing exists strongly enough to behave as if it does.

Mechanically, you create a lifeless representation that implies functionality: a video, mock interface, scripted flow, or walkthrough. The system cannot actually deliver the outcome. You observe whether users mentally adopt it anyway.

The failure monster you are tempting out is weak belief. People find the idea interesting, even if they know it is not yet real.

The sign that matters here is users asking real questions, imagining use, or planning behaviour around something that does not yet work.

Dropbox’s explainer video.

Before Dropbox built syncing infrastructure, they released a short demo video showing files syncing effortlessly across devices. Users signed up in large numbers and shared the video. The product was not functional. The behaviour showed belief strong enough to change expectations and intent.

This remains relevant because the technique is about belief, not technology. AI makes Pinocchio easier to do, and more dangerous to misread.

Mechanical Turk

Mechanical Turk pretotypes test outcome value.

They ask whether the result itself matters enough, even when delivery is manual, slow, and visible.

Mechanically, humans openly deliver the outcome. There is no illusion of automation. The experience may be clumsy. That is the point.

The failure monster you are tempting out is that the outcome does not matter enough to earn repeat behaviour.

The sign that matters is people tolerating imperfect delivery and still coming back.

Scale AI’s early data-labeling service.

Before building large-scale automation, Scale relied heavily on human labelers delivering training data to customers. Clients were buying outcomes, not systems. The test was whether clean labeled data mattered enough to justify cost and friction before automation improved margins.

This fits Mechanical Turk because value was tested before efficiency.

Wizard of Oz

Wizard of Oz pretotypes test experience fit.

They ask whether the experience works when it feels real, even if the system behind it is not yet autonomous.

Mechanically, you create the illusion of a complete system while humans secretly operate or assist it. From the user’s perspective, the system appears to work end-to-end.

The failure monster you are tempting out is experience breakdown over time: trust erosion, friction, confusion, or abandonment.

The sign that matters is repeat use under the belief that the system is autonomous.

Amazon Go early cashierless stores.

In the early rollout, the stores delivered the “just walk out” experience, but humans assisted and corrected edge cases behind the scenes. Customers behaved as if the system was fully automated. Amazon observed whether people trusted the experience enough to shop normally and return.

This was a Wizard of Oz because the experience was tested before full autonomy.

Provincial

Provincial pretotypes test real-world friction.

They ask whether the idea survives in one specific, constrained environment.

Mechanically, you launch the real thing, but only in a deliberately small context: one city, one customer segment, one use case.

The failure monster you are tempting out is the real world killing the idea.

The sign that matters is an ‘it’ that survives.

Monzo’s UK-only rollout before international expansion.

Monzo launched as a fully real bank, but only in one geography with a controlled user base. The provincial constraint forced the model to prove itself under real regulatory and behavioural conditions before scaling.

Re-label

Re-label pretotypes test meaning.

They ask whether the idea is right but the promise is wrong.

Mechanically, you change language, framing, or positioning without changing the underlying thing.

The failure monster you are tempting out is demand being driven by meaning rather than the performance.

The sign that matters is a behavioural shift caused purely by wording.

Slack reframing from “IRC replacement” to “where work happens”.

The product did not materially change. The framing did. Adoption broadened dramatically once the promise shifted from a technical tool to a work coordination space.

Pretend-to-Own

Pretend-to-Own pretotypes test commitment.

They ask whether your concept is strong enough to tempt your users to put some skin in the game..

Mechanically, users are asked to behave as if they already own or rely on the thing: deposits, reservations, paid pilots, data handover, calendar locks.

The failure monster you are tempting out is interest collapsing once something real is at stake.

The sign that matters is follow-through.

Tesla Model 3 reservations.

Customers placed refundable deposits years before delivery. The test was not attention. It was willingness to commit money early.

This separated excitement from seriousness at scale.

Stripped Tease MVP

Stripped Tease MVPs test repeat behaviour.

They ask whether the behaviour loops once everything else is removed.

Mechanically, you strip the experience down to the single behaviour that matters most.

The failure monster you are tempting out is one-time novelty.

The signal that matters is second and third use without prompting.

Instagram’s pivot from Burbn.

The team removed everything except photo capture, filters, and sharing. The stripped experience either created a loop or died. It lived.

This technique remains essential in tech and AI products that otherwise sprawl too early.

Kill, change, or build

Pretotyping has only three valid outcomes. Kill the ‘it’. Change the ‘it’. Or earn the right to build it properly.

Most ‘its’ should die here. Pretotyping works by creating xemes early enough to avoid big costs.

The process acknowledges that the failure monster can’t be killed, it can only be tricked. Feed it small meals early so it never gets the chance to eat the whole project.

Define the ‘it’ in one sentence. Choose the smallest bait that could expose the monster. Decide what action counts before you run it. Run it for a fixed number of real opportunities. Decide kill, change, or build.

Bibliography & Further Reading

The Xeme Framework draws on Alberto Savoia’s seminal work on pretotyping. You can learn more about his book The Right It here

You can also watch him explain the concept in his own words here

Check out some of my other Frameworks on the Fast Frameworks Substack:

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

The Xeme Framework

Aditya Sehgal — Sat, 17 Jan 2026 20:39:12 GMT

TL;DR — The Xeme Framework

Just as biology evolves through genes, and ideas spread through memes, in the Xeme Framework, I introduce the xeme as the way useful learning propagates in markets and organisations.

A xeme is the smallest transferable unit of learning created by a real-world test and embedded into a system.

Competitive advantage comes from xeme velocity: how fast a system can create, kill, propagate, and inherit learning.

Systems that embed xemes into products, processes, and defaults compound progress. Systems that don’t end up relearning the same lessons again and again.

This framework is about how to harness xemes to drive hypergrowth.

China is used as a live case study, not because it is unique, but because its competitive environment makes xeme dynamics visible at full speed.

For years, China was where Western companies went to grow. Now it is where they are being outcompeted. Chinese companies have dominated at home across sectors and are rapidly expanding abroad.

In 2024, Porsche’s sales in China fell by roughly 28 percent. BMW’s dropped by more than 13 percent, pulling down global performance because China had been one of its most important profit pools. General Motors took a multi-billion-dollar write-down on its China business after sustained losses. Starbucks reported a double-digit fall in same-store sales in China while local competitors continued to expand. H&M saw sales collapse following consumer backlash and has since reduced visibility and ambition in the market.

These outcomes are often explained through geopolitics, regulation, or cultural missteps. All of those contribute. But even when you account for them, a deeper, more general pattern remains.

The core issue is structural.

Chinese markets generate learning at a rate that overwhelms old school organisations built for planning, control, and linear improvement. Companies that survive do so because they adapt faster, not because they forecast better.

China’s advantage is not just creativity, scale, or cheap labour. It is learning speed under pressure.

The Xeme Framework

The Xeme Framework shows how to build organisations that evolve and scale quickly

From genes to memes to xemes

In biology, evolution works because genes carry learning forward. Variation is tested by the environment. What fits survives through inheritance.

In culture, Richard Dawkins introduced the idea of the meme: a unit of cultural transmission that spreads through imitation. Memes compete for attention. The ones that resonate propagate.

Markets evolve in a similar way, but through execution rather than imitation.

A xeme is a term I constructed to describe this process. It is deliberately new, because existing business language blurs the difference between having an insight and changing how a system behaves.

A xeme is produced when something is executed in the market and results in a clear outcome.

If the outcome is positive, the xeme is retained and carried forward.

If the outcome is negative, the xeme is killed and disappears.

Xemes are not ideas, opinions, intentions, or post-mortems. They are earned through action and confirmed by behaviour.

The pace at which a system evolves depends on how many xemes it can create, select, kill, propagate, and inherit per unit time.

The Wheel as a xeme

An early xeme may have emerged when a human tested a round object and noticed that it rolled. The experiment worked. That learning did not remain an isolated observation. It was embedded.

First in a simple cart. Then in wheels. Later in pulleys, mills, gears, and transport systems.

Like genes in a biological system, xemes persist through inheritance. They are reused, recombined, and refined across generations of products and ideas.

You can trace the wheel in a modern car back to that early xeme: the retained learning that a circular form enables efficient rolling. Materials and tolerances have changed. The underlying learning has not.

This is why iteration alone is not enough.

Evolution comes from retaining what works and building on it.

A system that runs many tests but forgets the learning generates activity without progress.

A system that runs few tests learns slowly.

A system that runs many tests and reliably embeds successful xemes evolves rapidly.

The importance of iterations vs the importance of time

We think compound interest is a function of time. Not so.

Iterations beat time. Compounding is not primarily a function of elapsed time. It is a function of how many iterations you can run in an iterated game. If you can create more iterations in a shorter period, you compound faster. A xeme is the compounding element of the system. The more winning xemes you generate and embed, the faster the system compounds.

How xeme systems actually evolve

All Darwinian systems share the same core logic. What differs is how efficiently each stage runs. To avoid ambiguity, it helps to separate the mechanics explicitly.

Creation (variation)

New candidates are introduced: features, prices, messages, bundles, processes, architectures, supply configurations.

Selection

The environment decides what survives. In markets, this shows up through behaviour and economics: conversion, repeat, cost, speed, willingness to pay, churn, substitution.

Killing

Failed variants must be removed.

In slow organisations, they linger, accumulate defenders, and consume attention.

High-xeme systems kill quickly and quietly, keeping the system clean.

Propagation

Winning xemes spread.

They move from one product to the next, from one team to another, from one category to adjacent ones.

Inheritance

Winning xemes are encoded into defaults, constraints, rules, systems, and architectures.

Propagation distributes winning variants.

Inheritance locks them into how the system now behaves.

Many organisations are weak at this final step. They generate activity and even insight, but they fail to encode learning. As a result, the same lessons are rediscovered repeatedly by different teams at different times and places.

Here’s what inheritance looks like in practice. A winning creative pattern stops being a one-off “great ad” and becomes a template enforced by tooling and default briefs. A winning price-pack architecture becomes the default SKU ladder reused across categories. A winning supply response becomes an MOQ and lead-time rule encoded in procurement systems.

In China advantage comes not just from higher test volume. It is faster killing, propagation, and inheritance of winning xemes that drives winning advantage.

What qualifies as a xeme

To prevent “xeme” collapsing into “insight”, a learning must meet all five conditions.

Until learning is embedded, it does not alter future behaviour.

A practical proxy for xeme velocity is:

Embedded decisions per week or month that survive a real-world test and show measurable impact above a defined threshold (or survive a holdout test).

This avoids confusing motion with progress.

Why China produces xemes so fast

China’s market combines three conditions that dramatically accelerate the xeme lifecycle.

First, experimentation is cheap. Launching a new product, format, bundle, or message rarely requires long planning cycles or heavy approval structures.

Second, feedback is immediate. Platforms, mobile payments, logistics, and social commerce expose real demand in near real time.

Third, selection pressure is intense. Weak ideas lose quickly to better ones. There is little protection for incumbents and limited tolerance for underperformance.

Together, these conditions turn the market into a continuous xeme engine.

How xeme velocity shows up in China

China does not win because it has a different playbook.

It wins because xeme velocity shows up clearly and relentlessly across five domains.

Each domain touches multiple stages of the xeme lifecycle, but each one accelerates a different constraint. Together, they form a system that learns faster than most Western organisations are built to handle.

1. Speed as strategy (xeme creation)

Speed matters because it determines how many xemes a system can generate.

Products are launched into live markets before they are finished. The market itself becomes the lab.

Each launch is a test. Each test either creates a xeme or kills one. Failure stays cheap. Delay becomes expensive.

Luckin Coffee illustrates this clearly. New drinks are launched continuously. Real transaction data decides the outcome fast. Losers disappear within days. Winners are rolled out across thousands of stores almost immediately.

This is not chaos or lack of discipline. It is industrial-scale xeme creation.

Learning comes from volume, speed, and ruthless pruning.

2. Platform-native operating models (xeme selection)

In the traditional AIDA funnel, learning is fragmented.

Awareness is built through mass media
Interest is captured through search
Desire is shaped through social
Action happens in retail or checkout

Each stage sits on a different platform, owned by different teams, measured by different metrics. Drop-offs happen between stages. When a consumer disappears, the system rarely knows why. When something works, the learning arrives late and partial. The signal weakens at every handoff.

A xeme needs a tight loop: variant → test → metric → decision.

Fragmented funnels weaken that loop. There is too much friction and too much loss of information between stages.

In China’s social commerce model, this fragmentation collapses.

A single 15-second shoppable video on Douyin can generate awareness, interest, desire, and action in one continuous interaction. The consumer sees, wants, and buys without leaving the environment.

Live commerce driven by shoppable video is already the core driver of growth and advantage in Chinese ecommerce. This is no longer experimental. It is the operating norm. You can identify winners and losers in Chinese consumer products simply by asking who has adapted to this reality and who has not.

The outcome is immediate. The signal stays clean. A xeme is created in real time.

Platforms in this world act as selection engines, not marketing channels.

3. Value engineering (xeme discovery under cost pressure)

Chinese firms do not compete by stripping features or racing to the bottom on price.

They compete by delivering the best possible product at a price the consumer is willing to pay, without compromising on either dimension.

They start with a price that clearly works in the market and engineer backwards. This forces trade-offs into the open: which features stay, which are removed, which components are redesigned, which suppliers are bypassed, which steps are automated.

Each trade-off is tested. Each test produces a xeme about what customers value and what they will pay for.

Some selection signals here are fast. Others arrive through returns, service rates, and repeat behaviour. Over time, these value xemes accumulate into deep advantage.

A concrete example is the Xiaomi YU7. In China, it delivers stronger features, functionality, and perceived quality than the Tesla Model Y, at a price closer to £25–30k. My Model Y in the UK costs around £53k.

Tesla has been forced to cut prices sharply in China to stay competitive. Xiaomi is still winning on features and quality. That is not branding. That is xeme-driven value engineering: fix the acceptable price, then iterate relentlessly until the product meets or exceeds expectations at that price.

4. Cultural fluency through fast testing (xeme creation, selection, and killing in meaning)

Culture behaves like an environment too.

In China, cultural meaning is not argued about in advance. It is tested in public, at speed.

Brands put out multiple cultural expressions quickly: tone, colour, language, symbolism, references, formats. Most don’t land. Some do. Each outcome produces a xeme about what feels current, authentic, or aspirational to a specific audience at that moment.

A good example is Li-Ning.

Li-Ning’s resurgence is often described through Guochao, a term that broadly means a “national trend”: modern consumer brands reworking Chinese cultural symbols and pride into contemporary, globally legible products. This did not come from a single rebrand or a flash of creative genius.

It emerged through repeated testing at the intersection of sport, streetwear, and Chinese cultural references. Early attempts varied widely in tone and execution. Many failed. A small number resonated strongly. Those winning cultural expressions were retained and reused across products, campaigns, and collaborations.

Over time, these retained cultural xemes formed a coherent foundation. New designs built on what had already proven to work. Demand accelerated. Copycats appeared. Li-Ning’s advantage came from iterating faster on the same cultural base, not from constantly searching for something new.

Cultural fluency here is not taste or intuition.

It is the ability to generate, kill, and inherit cultural learning faster than competitors.

5. Networked supply as a learning system (xeme propagation and inheritance)

Supply determines whether winning xemes can propagate.

This is where SHEIN is the clearest example.

SHEIN does not treat supply as a static cost centre. It treats it as a learning system. Designs are launched in micro-batches, sometimes as small as a few dozen units. Real demand data determines what survives. Losers are killed quietly. Winners are scaled rapidly.

Each production decision produces a xeme: about price elasticity, fit, fabric choice, lead time, quality tolerance, and trend decay. When a design xeme proves itself, SHEIN’s software-driven supplier network allows it to propagate across factories, regions, and categories almost immediately.

Crucially, these learnings do not live in decks or presentations. They are embedded into procurement rules, MOQ thresholds, lead-time assumptions, and supplier scoring systems. That is inheritance.

This is why networked supply matters. Without it, even the best product or cultural xeme remains local. With it, winning xemes become systemic advantage.

Xemes across domains

A xeme is the same idea everywhere. What changes is the forms in which it shows up across domains.

This is why the xeme is a useful abstraction. It explains learning across products, culture, technology, operations, and strategy using one underlying mechanism.

AI as a xeme accelerator

AI does not create intelligence. It accelerates evolution. It increases xeme velocity in three ways.

Creation

AI lowers the cost of generating variants: designs, copy, prices, code, workflows.

Selection

AI detects patterns earlier in messy data, reducing ambiguity about what is working.

Inheritance

AI makes retention easier. Winning learning can be encoded into models, workflows, recommendations, QA checks, pricing systems, and defaults.

This is why high-growth companies in very different places share the same underlying strength. NVIDIA’s advantage reflects decades of retained architectural and ecosystem xemes. Netflix’s advantage reflects continuous experimentation with strong institutional memory. OpenAI’s advantage reflects rapid deployment, wide feedback, and fast embedding of what works.

Xemes are not nationalistic.

They thrive wherever environments allow cheap experimentation, fast feedback, and reliable retention.

Structuring organisations to leverage xeme driven growth

Under sustained selection pressure, organisations reorganise themselves.

Why xeme velocity compounds

This is why performance often looks chaotic early on, then accelerates suddenly. The S-curve is accumulated xemes crossing a threshold. Remember, hyper growth is driven why iterations, not time. Compress maximum iterations in the minimum time.

Im my hyper growth framework I describe how linear sequential improvements in four dimensions - traffic, conversion, basket size and repeat rate create a multiplicative effect whose mathematical product results in a a exponential growth trend. Put simply, an organisation which creates an accelerating list of xemes on those 4 dimensions will scale exponentially.

So what should you do?

Make xemes visible. Track embedded decisions that survived market tests.
Lower the cost of failure and raise the cost of delay.
Collapse feedback loops wherever possible.
Build systems that retain and reuse learning.
Treat value and culture as testable systems, not declarations.
Structure the organisation in a way that creates and captures memes in ways where the result is multiplicative, not just additive.

The inevitable outcome

Ive always felt it is a superpower to see whats happening in China (and Silicon Valley) today - because that enables you to predict the future as other people and places do the same things a few years later. The Chinese model where speed of iteration is a core competitive advantage along with a data driven networked ecosystem will be the default model for winning companies over the next decade.

This is not about China winning. It is about environments selecting.

As competitive pressure increases and AI lowers the cost of variation, organisations that generate and retain xemes faster will evolve faster. Those built for slower learning loops will fall behind until the gap becomes structural.

China shows this future clearly because selection pressure is high and feedback loops are tight. The same logic spreads wherever those conditions appear.

Others will follow, not out of admiration, but because the environment leaves them no alternative.

Bibliography & Further Reading

The Xeme Framework draws on work across evolutionary biology, economics, platform theory, and real-world operating systems. The sources below are useful lenses for understanding how learning compounds through variation, selection, and inheritance.

Evolution, learning, and selection

On the Origin of Species — Charles Darwin https://www.gutenberg.org/ebooks/1228
The Selfish Gene — Richard Dawkins https://en.wikipedia.org/wiki/The_Selfish_Gene
An Evolutionary Theory of Economic Change — Richard Nelson & Sidney Winter
https://press.princeton.edu/books/paperback/9780674272286/an-evolutionary-theory-of-economic-change

Experimentation, fast learning, and institutional memory

The Lean Startup — Eric Ries
https://theleanstartup.com/book
Experimentation Works — Stefan Thomke
https://www.hbs.edu/faculty/Pages/profile.aspx?facId=6568
Netflix Tech Blog
https://netflixtechblog.com

China-specific competitive dynamics

AI Superpowers — Kai-Fu Lee
https://www.amazon.com/AI-Superpowers-China-Silicon-Valley/dp/132854639X
Rest of World — SHEIN supply chain analysis
https://restofworld.org/2022/shein-fast-fashion-supply-chain/
McKinsey — China consumer and retail insights
https://www.mckinsey.com/featured-insights/china

Social commerce and funnel collapse

Bain & Company — China social commerce
https://www.bain.com/insights/topics/china-social-commerce/
Financial Times — Livestream commerce in China
https://www.ft.com/content/6a1e2a9e-2b78-4e3a-9f93-ec7a4f87a3c5

Value engineering and backward design

Competing Against Luck — Clayton Christensen
https://www.amazon.com/Competing-Against-Luck-Innovation-Customer/dp/0062435612
Toyota — Toyota Production System
https://www.toyota-global.com/company/vision-and-philosophy/production-system/

AI as an accelerator of learning systems

OpenAI — Research and system cards
https://openai.com/research
NVIDIA — CUDA & developer ecosystem
https://developer.nvidia.com/cuda-zone
Landing AI — Data-centric AI (Andrew Ng)
https://landing.ai/data-centric-ai/

The Value Creation Framework

Aditya Sehgal — Fri, 16 Jan 2026 03:38:57 GMT

TLDR:

The Value Creation Framework separates three distinct drivers that are often conflated:

Value = Impact × Translation × Multiple

Impact is execution – how much real work gets done.

Translation is economics – how much of that work becomes surplus.

The multiple reflects trust – how far into the future others are willing to capitalise that surplus.

Different systems optimise different parts of this equation. China is maximising impact and reinvesting surplus to build capability and scale. The US is maximising multiples by owning choke points and concentrating trust in a small set of firms. The framework is universal: it applies to countries, companies, and individuals. Long-term value is created when execution, economics, and trust reinforce each other.

Folks following western media feel that China has recently had a difficult run. Property has stayed weak, confidence has taken longer to return, and the outside narrative is still stuck on slowdown and decoupling.

Against that backdrop, the headline number from 2025 lands with a thud: China recorded a $1.19T trade surplus — the largest surplus any country has ever reported — up from roughly $990B in 2024, nearly a 20% jump.

That is large enough to force a rethink. It sits awkwardly beside the mood music, and beside the way many people talk about China today.

This becomes easier to understand once you stop compressing three different things into one: real-world activity, economic surplus, and financial value. They overlap, and they often move together, but they behave differently and respond to different incentives. A system can do a lot of work in the real world and still be priced modestly in markets. A system can also be priced aggressively long before its full real-world impact shows up. Most confusion comes from treating them as the same thing.

The framework

The simplest way I’ve found to keep this straight is a three-part lens:

Value = Impact × Translation × Multiple

Impact: how much real activity happens.

Translation: how much of that turns into surplus.

Multiple: how the market prices that surplus.

These multiply rather than add. A weak link in any one of the three pulls the outcome down, even when the other two look strong. This is as true for countries as it is for companies.

Getting all three right over time is how compounding happens.

A 1X improvement in each leads to an 8X increase in value.

What this lens makes clear is that value outcomes are not linear and they are not intuitive. Two economies can generate similar levels of activity and end up with very different financial results because they sit in different places on these three dimensions. The matrix that follows simply makes this visible.

One axis shows economic force (the number from today) – impact multiplied by translation – and the other shows the multiple (driven by the story from tomorrow), which captures how far into the future markets are willing to capitalise that surplus.

‘Every market valuation is a number from today multiplied by a story about tomorrow’ - Morgan Housel

Where a country or company sits on the chart explains why it is being rewarded or discounted, without any judgement about whether the underlying system is “good” or “bad”. It also makes clear that this is not static. Movement on the grid over time – improving translation, building trust, or changing how surplus is captured – is what ultimately drives long-term value creation.

Impact

China was still the biggest incremental engine in the real economy in 2025. Roughly $0.9T of added output, versus about $0.6T for the US, $0.3T for the EU, and $0.24T for India. [approx numbers based on projections]

Now look at what equity markets did over the same year (year-end 2025 market cap). The US finished at $71.0T (+$10.0T). China + Hong Kong finished at $23.6T (+$4.8T). The EU finished at $12.1T (+$2.0T). India finished at $5.3T (+$0.7T).

Put bluntly: the US turned roughly $1 of GDP growth into ~$16.6 of equity value. China turned roughly $1 into ~$5.3. The gap is real and persistent.

That gap doesn’t require a story about who’s ‘right.’ It requires a description of how impact turns into surplus, and how surplus gets priced.

Translation

The ‘China is slowing’ story looks weaker once you see where the surplus is coming from. Some parts still carry friction - property and household sentiment. But other parts are compounding.

The trade engine is the easiest place to see it.

China absorbed a large, visible negative shock in exports to the US (–$95B)/-20%, but simultaneously expanded exports across ASEAN, Africa, the Middle East, Latin America, and Europe by +$180B, leaving a net export increase of +$85B.That is what “decoupling” looked like in the data..

The trade balance makes the point even clearer. The surplus widened across multiple key markets, and China improved its balance with most major partners - by more, in aggregate, than the US did.

These are big numbers because they are structural, not cyclical.

China has become the default industrial partner of choice for a widening set of markets, especially where infrastructure build-out is still happening and price sensitivity is high. The US decline is real. It’s also a smaller share of the total story than it used to be. [For a deep dive down this rabbit hole, read my previous post on The China Struggle]

The product mix tells the deeper story of translation. The export categories that grew in 2025 skewed toward capital-dense, system-building goods. And at the same time, China reduced its share of low-priced consumer goods by pushing more of that manufacturing to other countries.

This shift isn’t a minor swing. It’s a re-weighting away from labour-intensive, low-margin categories and toward categories that sit closer to long-term industrial capability. This is what you would expect an innovative, hi tech superpower to export.

China is becoming the world leader in the key technologies of the future - in many cases aided by US withdrawal (Clean energy, battery technology, EVs are only a few examples)

EVs make it legible. China’s incremental exports in Autos & EVs alone (outside the US) more than offset the entire hit from lower US imports across categories.

China exported about 6.7M vehicles in 2025, including about 2.3M EVs and plug-in hybrids.

The exact split is less important than the shape: the export base is now broad, and the engine has routing options. ASEAN, LATAM and Africa + Middle East are all not that far from the volumes sold into Europe!

China isn’t just selling cars. It’s exporting an entire system – vehicles, batteries, charging hardware, grid equipment, financing, and service networks. That builds adoption and capability at scale, even if it keeps margins thin.

That makes the trade system more resilient. But it also keeps margins under pressure, because scaling fast across price-sensitive markets is a brutal game. The country gets capability. Shareholders get a smaller slice of surplus in the near term..

Now mirror the logic for the US, because symmetry keeps the analysis honest. The overall US trade deficit stayed very large in 2025, around $1.02T depending on the definition.

Within that, the US balance with China improved by about $25B as imports from China fell even more than exports to China. In the same year the combined balance with ASEAN and India worsened by $58B, which is the leakage effect: low end consumer goods moved through connector economies, aided by aggressive Chinese offloading and offshoring. The US surplus with the EU and the UK grew by about $33B, driven mainly by LNG, Oil and aerospace. That is a different kind of trade win. It sits in high scale, contract-backed lanes, and it converts into surplus more reliably than broad consumer goods import substitution.

Energy as a macro underwriter

The “subscription to Western stability” line deserves to be more than a metaphor. It is a description of contract structure and the way that structure feeds into discount rates.

US LNG exports into Europe are not primarily one-off spot trades. The backbone is long-dated contracting, often 10–20 years, dollar-linked pricing, and in many cases take-or-pay clauses. That creates committed volumes and more stable spreads than many commodity exposures. In valuation terms, this changes the distribution of macro outcomes. It reduces the frequency and severity of energy shock scenarios, it damps inflation volatility, and it reduces tail risks that would otherwise sit in the discount rate.

That matters most for long-duration assets, which is exactly what the big tech platforms are. When a large part of a company’s value sits far into the future, small changes in discount rates have outsized effects on present value. A world with more stable energy and fewer inflation spikes supports higher sustained multiples for companies whose value is driven by long-dated cash flows. This is why the energy corridor can matter a lot for equity markets without energy stocks being the main driver of equity gains.

Multiple and concentration: why value ends up where it does

China clearly created more absolute GDP than the US in 2025. It also exported more than the US, including in high-tech categories. On any simple measure of output, production, or physical economic contribution, China did more work.

So the obvious question follows: did investors in Chinese equities get rewarded proportionately to that growth, relative to the US?

Did Chinese investors get comparatively rewarded for the work done by their companies.

The answer is no.

In 2025, roughly 53% of total US equity market gains came from just ten companies. In China plus Hong Kong, the equivalent figure was closer to 29%, with the remainder spread across a much broader set of firms.

Those numbers describe concentration, but they don’t yet explain it.

Concentration is not an independent force. It is the visible outcome of sustained differences in valuation multiples, applied repeatedly over time.

When a small number of companies trade at materially higher multiples than the rest of the market, every incremental dollar of profit they generate adds disproportionately more market value than the same dollar generated elsewhere. If that condition persists year after year, the effect compounds. Market capitalisation drifts toward those firms even if growth exists across the system.

Once that divergence begins, it becomes path-dependent.

Index construction, passive flows, benchmark constraints, and risk models reinforce the outcome mechanically. As a company’s weight in an index rises, more capital is forced into it. That incremental capital reduces volatility, which supports higher multiples. Higher multiples then attract more discretionary capital, which further increases index weight. Over time, leadership becomes self-reinforcing.

Buybacks amplify the process further.

In the US, large technology firms have spent years converting operating cash flow into per-share value. As profits rise, share counts fall. Earnings per share grow faster than operating profit. Multiples are applied to a shrinking base. Market capitalisation accelerates again.

None of this requires heroic assumptions. It is arithmetic interacting with market structure.

This is why concentration is not a distortion and not a mystery. It is the footprint left behind when high multiples, buybacks, and index mechanics interact over long periods of time.

What still needs explaining is why the market is willing to keep applying those high multiples to the same set of companies, cycle after cycle — and why similar-looking businesses elsewhere do not receive the same treatment.

What allows these multiples to persist: trust

One lesson I’ve learned over time is that most companies are highly disciplined about managing the first two parts of the value chain, and far less deliberate about nurturing the third.

Inside organisations, almost all the focus sits on impact and translation.

Teams work relentlessly on driving sales, expanding market share, shipping more units, entering new markets, and building scale. Then they turn to margins, cost control, operating leverage, and cash conversion. All of that is necessary. None of it is optional. If a business is weak on impact or translation, there is no value to talk about.

But that is not where the disproportionate reward comes from.

The disproportionate reward, especially in public markets, shows up in the multiple.

And the multiple is not driven by last quarter’s numbers. It is driven by trust that the company will deliver financial performance consistently over a very long period of time.

This is where the Value Translation Framework links directly to the Trust Framework. The multiple is not a mysterious market artefact. It is the market’s shorthand for a judgement about trust.

In practice, that trust rests on three things.

The first is credibility and intimacy.

Does the company set sensible targets, and does it meet them? Markets place more weight on consistency than on ambition. Companies that repeatedly say what they will do and then do it earn the right to be extrapolated. Companies that over-promise, even if they occasionally outperform, teach the market to discount them.

The second is reliability.

This is about behaviour over time and across conditions. Can the company deliver through cycles, shocks, and uncertainty? Reliability is what allows investors to project today’s cash flows forward without constantly haircutting them. Without it, even strong current performance is treated as fragile.

The third is low self-interest, and this is the dimension most leadership teams underestimate.

Investors watch governance closely. They watch how capital is allocated, how incentives are structured, how management behaves when trade-offs appear, and whether decisions consistently favour long-term shareholder and stakeholder outcomes rather than short-term managerial gain.

Companies that are seen to do the right thing when it costs something build a deep reservoir of trust. Companies that don’t eventually see that trust erode, regardless of how attractive the numbers look.

Taken together, these three dimensions are what the market is really pricing when it assigns a multiple.

The multiple is not just a valuation ratio. It is a statement about how far into the future the market is willing to believe.

This is why two companies with similar growth rates and similar margins can trade at very different valuations for long periods of time. One has earned trust across credibility, reliability, and low self-interest. The other has not.

Seen through this lens, the framework becomes a practical diagnostic rather than an abstract explanation. The key question shifts to where value is leaking.

Is impact strong but translation weak because pricing power or cost structure is the constraint?
Is translation strong but the multiple capped because the market does not trust durability?
Or is the underlying business sound, but governance, capital allocation, or communication are undermining belief?

Pricing power, cost structure, capital intensity, buybacks, concentration, and narrative consistency all sit in different parts of the chain. A CEO’s job is not simply to push harder on growth or squeeze margins, but to understand which lever actually moves value at the margin at that moment in time.

Most leadership teams are trained to manage impact and translation. Far fewer are trained to manage trust deliberately. Yet in public markets, trust is often the difference between a good business and a great investment.

Why this trust dynamic penalises Chinese multiples — by design

This trust lens also explains a large part of the persistent multiple gap between US and Chinese companies, including in technology.

Many Chinese companies are credible. Many are reliable. Many execute extremely well. The gap is not capability.

The difference lies in how value is intentionally distributed.

In China’s system, value created by firms is expected to serve a wider set of stakeholders: consumers, workers, suppliers, local governments, and national priorities. Price competition has historically been tolerated - and often encouraged - as a way of accelerating adoption, lowering costs, and building industrial depth.

This is where the concept of “involution” (内卷) matters.

For a long period, Chinese firms competed in ways that destroyed margins without creating incremental capability. That was recognised as a problem. Over the last 18–24 months, policy has shifted toward discouraging irrational price wars, encouraging consolidation, and emphasising quality, standards, and technological depth.

But that shift is recent, and the legacy of involution still shapes how markets view Chinese firms.

From an investor’s perspective, the concern is not that profits will not be made. It is that surplus capture is less predictable.

When value is routinely passed through to consumers via lower prices, reinvested into capacity, or absorbed into broader system objectives, the horizon over which equity cash flows can be confidently capitalised shortens.

In Trust Framework terms, many global investors struggle less with credibility and reliability in Chinese companies than with low self-interest, as they define it.

The concern is not malfeasance. It is uncertainty about how consistently minority shareholder interests will be prioritised when they conflict with system-level objectives.

That uncertainty caps multiples even when operating performance converges.

This does not mean Chinese companies cannot earn higher valuations. It means they face a higher bar. They must demonstrate not just performance, but alignment, transparency, and capital discipline over long periods of time, across changing policy and geopolitical conditions.

In addition, multiples are not helped by the increasing media noise and social chatter in the west in creating an us-vs them, decoupling narrative,

That is why the multiple gap persists even when growth rates, margins, and technology leadership converge.

The company lens: the same framework in miniature

At company level, the same logic becomes even clearer.

Impact is revenue growth and units shipped.

Translation is conversion into EBITDA and cash.

The multiple reflects trust and the expected durability of surplus capture.

Nvidia makes the logic painfully legible.

Over the period discussed, Nvidia’s revenue increased by about $40–45B. Over the same period, its market capitalisation increased by roughly $1.5–2.0T.

That is an extraordinary conversion. Every $1 of revenue drove $16 in market value.

Translation is strong because Nvidia sits at a scarcity choke point, with pricing power, capital-light economics, and operating leverage. Capture is extreme because a large share of global AI value accrues to it before value reaches downstream users.

Markets can see that chain clearly, and they capitalise it far into the future.

BYD is the counterexample, not because it underperforms, but because it performs a different role.

BYD sold about 3M vehicles in 2025. Revenue increased by roughly $25–30B. Real-world impact was enormous: transport electrified, costs lowered, capacity built across China, ASEAN, Latin America, the Middle East, and Africa.

But BYD’s market capitalisation increased by only $80–120B. 1$ revenue drove only $4 in value

That gap is not a failure of impact. It reflects translation economics shaped by intense price competition, reinvestment into capacity, and the role the company plays in building system capability rather than extracting rents.It does not adequately consider the long term profit pool and the higher share that BYD will eventually capture of the global auto profit pool because of scale.

Value is created and used widely before it ever concentrates in equity value.

That is not inefficiency. It is design.

Value creation versus value recognition

China creates a lot of value that is used before it is priced. Roads, power systems, industrial ecosystems, EV fleets, charging networks, manufacturing capability, logistics reliability, and learning curves show up in daily life and national capability long before markets fully recognise them. The real-world value arrives early. Recognition is often delayed.

The US often prices value before it is fully realised. Expectations about future dominance, future margins, and future optionality are capitalised early. Sometimes that future is delivered, sometimes it takes longer, sometimes it changes shape, and sometimes it proves fragile. The ordering still matters. One system builds, then gets priced. The other system gets priced, then builds.

Using the framework as a diagnostic

Companies:

Where I’ve found this framework most useful is as a diagnostic for CEOs, because it surfaces a gap that many leadership teams sense but don’t always name.

Most companies are already quite good at managing the first two parts of the chain. Impact shows up in sales growth, units shipped, customers added, and market share. Translation shows up in margins, cost discipline, operating leverage, and cash conversion. This is where operating reviews focus and where incentive plans are built. And it should be. If you’re not growing, or if growth doesn’t convert into profit and cash, there’s nothing to value.

But that still isn’t where the real payoff comes from.

The real payoff shows up in the multiple. And the multiple is not driven by last quarter’s numbers. It reflects whether the market believes the business can deliver financial performance consistently and defensibly over a long period of time.

The arithmetic makes this clear. A business earning £100m of sustainable annual profit on a multiple of 10 is worth £1bn. If nothing changes operationally, but the market becomes willing to value that same profit stream on a multiple of 11, the valuation rises to £1.1bn. One extra turn on the multiple has created £100m of value — roughly a full year of profit — without selling a single extra unit. In many capital-light businesses, a one-point move in the multiple can be worth five to ten times annual profit.

Seen this way, the diagnostic question becomes practical: where is value leaking?

Sometimes impact is strong but translation is weak because pricing power or cost structure is the constraint. Sometimes translation is strong but the multiple is capped because investors don’t trust how durable those profits are. And sometimes the business is sound, but capital allocation, governance, or communication quietly undermine belief. Running the business well mostly moves the first two. Being rewarded for it depends on the third.

The multiple rises when investors trust that the company will do what it says, do it repeatedly, and do it in a way that aligns management behaviour with long-term shareholder and stakeholder outcomes. Credibility, reliability, and low self-interest are not abstract virtues here; they determine how far into the future the market is willing to extrapolate today’s profits.

Running the business well gets you profit.

Earning trust determines how many times that profit is worth.

Governments:

A government can use it to make trade-offs explicit. Policies that maximise impact and capability often suppress translation in the short term. Policies that protect margins and capital returns can lift translation and multiples, at the cost of diffusion and affordability. The framework helps policy-makers choose which lever to bias at which stage of development, instead of pretending all objectives can be maximised simultaneously.

Investors:

An investor can use it to decide what kind of exposure they are buying. Some assets maximise impact but rely on changes in translation and multiple over time. Others maximise current translation and multiple but depend on discount rates and belief. Framing it this way turns asset allocation into a structured choice rather than a narrative fight.

In closing

Separating impact, translation, and the multiple forces you to be honest about where value is actually flowing, and where it’s simply being priced.

Once you see it that way, debates about whether China or the US is “winning” start to feel beside the point. What matters is understanding which system you’re operating in, which lever it prioritises, and whether the gap between use and recognition is widening or narrowing over time.

When I look at this through my own lens, what stands out is how intentional both systems are.

China’s system does what it is designed to do. It compounds scale, networks, and physical systems, and it deliberately reinvests surplus to build capability in industries that matter next. Value shows up first in use: in cheaper transport, more reliable logistics, denser supply chains, and faster learning curves. Financial recognition comes later, if at all, and only once capability is secure.

America’s system also does what it is designed to do. It prices the future early. It applies high multiples to long-duration cash flows, and then reinforces those valuations through buybacks, index mechanics, and a market structure built to convert operating success into per-share value as quickly as possible. Value shows up first in market capitalisation, often well before the full real-world impact is visible.

Neither approach feels accidental to me. They reflect different answers to the same underlying question: where should surplus go, and when?

One system pushes surplus back into the real economy to strengthen position and resilience. The other pulls surplus forward into financial markets to compound capital. The outcomes look very different on a screen, but they are coherent once you follow the chain.

That’s why I’m comfortable holding two ideas at the same time: that China is doing an extraordinary amount of real work in the world, and that the US is extraordinarily good at getting paid for the work it chooses to recognise early. The tension between those two is not a contradiction. It’s the point.

Footnotes, definitions, and scope

1. Market capitalisation figures

All market capitalisation figures cited in this article (for example, US ~$71.0T with a ~$10.0T year-on-year increase; China + Hong Kong ~$23.6T with a ~$4.8T increase; EU ~$12.1T with a ~$2.0T increase; India ~$5.3T with a ~$0.7T increase) are end-of-year 2025 approximations based on aggregated exchange-level and index-level data.

These figures are not intended to represent a single canonical series. Reasonable alternative definitions — such as Wilshire 5000 vs MSCI USA for the US, exchange-level totals vs index aggregates, different treatments of cross-listed securities, or FX conversion at spot vs average rates — will produce modest variations. None of those variations change the central observation: the US added materially more equity value than China+Hong Kong in 2025, and that value was far more concentrated.

2. Market-cap concentration (top-ten contribution)

Statements such as “approximately 53% of US equity gains came from the top ten companies” and “approximately 29% in China+Hong Kong” refer to estimated contribution to total market-cap change by the ten largest contributors within the relevant market universe.

The exact percentage varies depending on:

index or universe definition,
start and end dates,
FX treatment,
and whether the ranking is by absolute contribution or end-period size.

The argument does not rely on any specific percentage being exact. It relies on the robust pattern that US equity value creation in 2025 was heavily concentrated in a very small number of firms, while China’s was more broadly distributed.

3. Trade flows and surplus by region

Trade and surplus figures are drawn from China Customs Administration releases for calendar year 2025, supplemented by standard trade aggregations. Regional figures (ASEAN, Africa, EU, Middle East, Latin America, India) are groupings rather than individual bilateral claims.

As with all trade data, figures vary depending on:

FOB vs CIF treatment,
re-exports and trans-shipment,
exact country inclusion in regional groupings.

The analysis depends on direction, scale, and simultaneity rather than bilateral precision: a large decline in US-bound exports, offset by large increases across multiple other regions, resulting in a record overall surplus.

4. “Incremental real GDP in USD terms”

The phrase “incremental real GDP in USD terms” is used as a comparative heuristic, not a strict accounting construct. It refers to estimated year-on-year real GDP growth in local currency, translated into USD using average annual exchange rates for ease of comparison across countries.

This approach is acknowledged as imperfect. Its purpose is to convey relative scale — that China added more real economic output in 2025 than the US — not to produce a precise decomposition. The structural argument holds under alternative reasonable treatments.

5. EV export volumes and regional distribution

Vehicle and EV export volumes are based on industry reports and customs data for calendar year 2025, and include China-manufactured exports (ICE, EV, and PHEV as specified). Regional distributions are approximate and reflect destination markets rather than final end-use in cases of onward resale.

The analysis does not hinge on the exact unit count for any single region. It hinges on the structural shift toward a broad, multi-regional export base rather than dependence on any single market.

6. US trade deficit and energy corridor

References to the US trade deficit (~$1.0–1.1T) reflect goods and services combined, using commonly cited BEA/Census aggregates. Bilateral shifts (for example, narrowing with China and widening with ASEAN/India) are directional and illustrative.

The discussion of the US–EU energy corridor refers to the structure of LNG exports — long-dated contracts, dollar pricing, and take-or-pay provisions — rather than short-term spot pricing. The argument concerns how these structures influence macro volatility and discount rates, not the profitability of energy producers per se.

7. Company examples (Nvidia, BYD)

Company-level figures are illustrative and based on publicly reported financials and market-cap movements over the relevant period. The comparison is intended as an intuition pump, not as proof or a comprehensive peer analysis. The framework applies equally to other firms with similar characteristics.

8. On the Trust Framework

Where the article references the Trust Framework, it points to a separate framework describing how trust is driven by credibility, reliability, intimacy and perceived self-interest over time. The argument here does not depend on prior familiarity with that framework; it draws on the same underlying logic.

9. What this article is — and is not

This article is not a valuation model, a trading recommendation, or a claim of statistical precision. It is a structural explanation of why real-world impact, economic surplus, and financial value diverge across systems, and why those divergences persist.

The thesis stands or falls on the mechanisms described — impact, translation, and multiple — not on the exact decimal point of any individual figure.

Check out some of my other Frameworks on the Fast Frameworks Substack:

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

The Ladder and Wall Framework

Aditya Sehgal — Sun, 11 Jan 2026 23:08:19 GMT

The Ladder and Wall Framework is how I think about your progress and direction.

Capable people are good at climbing. Put a ladder in front of them and they will work out how to get up it. They climb roles, titles, pay bands, reputations. They climb expectations set by parents, managers, peers, and often by younger versions of themselves. The next rung is visible, the effort is familiar, and the climb feels productive. Momentum builds easily, which is why many people spend years climbing without stopping to ask where the ladder is actually taking them.

As time passes, the work keeps coming and the signals still look positive, but the range of future moves starts to narrow. Choices feel harder to make. From the outside it still looks like progress, so there is rarely a clear moment when you stop and reassess. You just keep going.

The real question is not whether you will succeed in the climb. If you are smart, disciplined, and driven, you probably will. The more important question is what the ladder is leaning against, and what sort of place you will arrive at when you reach the top.

In this framework, the ladder is what you are climbing. The wall is where that climb leads.

You can climb well, steadily, and for a long time, and still arrive somewhere you never consciously chose (and you dont’ like). That is the Ladder and Wall Trap.

Once you separate ladder from wall, the framework becomes practical. Four situations show up quickly, and most frustration around work and direction fits into one of them.

The Ladder-Wall Framework

Right ladder, right wall: Compounding

Effort builds on itself. Progress creates options rather than closing them. The work you do today makes tomorrow easier, and your sense of self stays aligned with where you are heading.

Right ladder, wrong wall: Comfort Trap

You perform well inside a structure that no longer fits. Skills deepen, rewards continue, and leaving gets harder over time. What feels like stability slowly turns into constraint.

Wrong ladder, right wall: Wrong Track

The environment still has energy, but your role within it does not. You feel under-used. Motivation fades. Opportunity is nearby, but you are not positioned to take advantage of it.

Wrong ladder, wrong wall: Going Nowhere

Effort goes in without much coming back. Work expands, energy drops, and years pass without much changing underneath. People here are often busy and exhausted at the same time.

The point of the matrix is to help you see where you are while there is still room to move.

What ladders and walls are

The ladder is your role, your skill set, your craft, your function, or the path you are currently on. It is the thing you get better at through repetition and effort. It rewards discipline and persistence, and it gives feedback quickly through results, recognition, and progress.

The wall sits behind all of this. It is the larger structure you are climbing into. The industry. The organisation. The place. The phase of life. The general direction things are moving over time. When the wall fits what you want, climbing makes sense and tends to pay off. When the wall weakens or moves away from you, climbing harder only increases how dependent you are on it.

A simple way to hold the framework is this. A good ladder gets you up. A good wall tells you whether starting again from the bottom is worth it.

The platforms between ladders

Between ladders, there are platforms.

When a climb ends, you do not move directly from one ladder to the next. You step off onto a platform first. That platform might come after a promotion, finishing a degree, shipping a major product, selling a company, or deciding to move cities. Something has finished. Something else has not yet begun.

While you are climbing a ladder, attention is narrow and mostly upward. You focus on the next rung, the next test of competence, the next signal that you are progressing. Platforms change that. They give you distance. From a platform, you can see where you are, what the climb actually gave you, and what it did not. You are no longer scrambling for footing, but you are also not yet committed to the next ascent. This is where judgment becomes possible again.

Most people do not linger on platforms. They step straight onto the next ladder because momentum feels easier than stopping. Over time, this turns platforms into missed moments of choice rather than moments of orientation.

Life as a sequence of ladders

Life does not unfold as one long climb. It unfolds as a sequence of ladders, connected by platforms.

You climb one ladder, reach a platform, then choose the next ladder. Often the next ladder starts lower than where you just were. That drop is part of changing direction. It is how movement happens.

Most people recognise this pattern when they look back. At the end of school, confidence is high. At university, everything resets. The same thing happens when entering work, changing roles, or shifting paths entirely. Height resets, direction continues.

Each ladder calls for a different mindset. Early on, learning dominates. Later, you value execution and judgment more.

Platforms ask you for honesty. They are where direction gets set, whether deliberately or by default.

A quick check

You can use the framework without turning it into an exercise.

If you are successful at what you are doing over the next five years, who will that success turn you into?

If things go well on this path, will you be closer to the person you want to be, or just better at playing this role?

If you imagine yourself standing on the platform five years from now, would you choose to climb the same ladder again?

The answers tend to surface quickly.

The role of AI

AI does not change the fact that people climb ladders. It changes how quickly walls move. Industries that looked stable for decades can tilt in a few years. Skills that compounded reliably can stall suddenly. The trap becomes easier to fall into because the ladder still works while the wall shifts while you were distracted climbing the ladder.

AI can help you climb faster. It takes friction out of learning, execution, and coordination. Used well, it compresses effort and shortens the distance between rungs.

It can also help you see the wall more clearly. AI is good at pattern detection across markets, skills, and trajectories that are too large or slow-moving for any one person to track. It can surface shifts in demand, emerging adjacencies, and declining paths long before they become obvious in day-to-day work.

AI also reduces the cost of exploration. It makes it easier to simulate paths, test assumptions, and look sideways from a platform without committing years of effort. You can try ideas, roles, directions, and combinations cheaply before you choose what to climb next.

The combination changes how platforms work. Walls that once moved slowly can tilt mid-climb. Ladders that still feel stable can quickly lose relevance. Platforms matter more because they are where speed gives way to judgment. They are where you decide whether to keep climbing, change ladders, or change direction altogether.

AI is most valuable when it is used at the platform, not just on the ladder. It helps you decide what to climb next, not only how quickly you can climb.

Choosing what to climb

The most important question in this framework is not whether you can climb a ladder. If you are reading this substack post, its very likely that you are interested in self development and have the capability to climb any ladder.

The more interesting question is what the view will be like when you get to the top. Will it be your promised land?

If being higher on this ladder makes it harder to move toward the life you actually want, pay attention. You can finish the climb, step onto the platform, and then move sideways into a different sequence of ladders. Height gives way to direction. Familiarity gives way to trajectory. That shift reflects judgment.

Smart people will always climb. What shapes the outcome is where the ladder is placed, what wall it leans against, and whether platforms are used deliberately.

Climb well. Take responsibility for what you are climbing into.

Direction compounds faster than effort.

Check out some of my other Frameworks on the Fast Frameworks Substack:

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro

The Funnel OS Framework

Aditya Sehgal — Sun, 04 Jan 2026 20:07:08 GMT

A funnel is any system that turns volume into an outcome through a sequence of gates. This framework focuses on optimising the Funnel ‘operating system’ and tuning it for exponential output growth by deliberately structuring the funnel and using AI in the right places

If you can’t name your gates, you can’t fix your conversion.

The same funnel structure shows up everywhere:

Hiring (applications → interviews → offers)
Sales (leads → calls → close)
Product (signup → activation → retention)
Content (impressions → clicks → subscribers)
Fundraising (intros → meetings → term sheets)

The Funnel OS Framework

Volume × Gates × Gate rates → Outcome

The wrong order leads to Gate Entropy (more steps, lower conversion, slower decisions). The right order leads to compounding growth.

Use AI to turbocharge each of the gates.

A Real Example: jobs.asgard.world

Jobs.asgard.world is an AI driven platform to help people find jobs.

At the top of the funnel, the platform has over 10 million fresh jobs around the world. Users upload their current CV. The platform extracts the users’ skills and experience. It then uses AI to match them to relevant jobs (in some cases hundreds of jobs), and the AI writes customised CVs and covering letters for each role – ensuring that the core keywords needed by the recruiters are covered, and the CVs are well structured and machine readable.

We ran a closed beta test with 100 users over 75 days. In this period, the platform produced:

42 users with interview calls
12 accepted roles

But outcomes were very different depending on how users ran the funnel:

55 users applied to between 0 and 2 roles. None of them got an interview call. 3 users applied to over 10 jobs but did not get any interview calls (they were in niche/senior roles). Of the 42 people who got interview calls, 5 people got over 5 calls each. And there was a clear difference in how these people ran the funnel.

The best-performing user got 12 offers. This person made 50 applications a day over 12 days (high volume of relevant roles at the top of the funnel). They used the optimised CV and covering letters, with further minor manual optimisations (low friction). And they were consistent in running the funnel everyday for many days

~50 applications/day for 12 days → ~600 applications
Over the next 40 days:
- 40 interviews
- 12 offers

The playbook:

Scale relevant volume [AI matching between candidate skills and job needs]
Improve conversion quality (ATS-readable CVs, job keywords, plain formatting).
Apply early. Timing is a gate.

The differentiators between people who got interview calls and those who didn’t were a) lack of application volume b) ignoring the customised CVs and using manual CVs which looked beautiful but weren’t machine readable or lacked keywords (causing a lower conversion rate) and c) lack of consistency.

1) Fill the funnel: maximise high‑intent volume

Most people buy cheap traffic and then wonder why conversion is terrible.

Pay for intent.

Example:

Facebook banner → Amazon purchase: 0.5–2%
Amazon banner → Amazon purchase: 15–30%

So you can pay 5× more per eyeball on Amazon and still win on ROAS.

2) Cut Gate Tax: minimum gates, maximum signal

Every gate is a tax.

Friction is the mechanism that jams the locks on the gates.

Every extra click, field, page load, or decision adds friction. Friction kills conversion.

Concrete examples of friction destroying conversion:

Checkout: forcing account creation before checkout.
Payments: hiding shipping costs until the final step.
Performance: slow page loads on mobile.
Forms: asking for “nice to have” fields (phone number, address, company size) too early.
Trust: unclear returns policy or missing delivery dates.
Timing: long delays between steps (e.g., “we’ll get back to you next week”).

Don’t anchor on a single universal number (it varies by context), but the direction is consistent: each extra step creates a material drop.

Every extra step costs:

conversion
time
trust

Keep only gates that do real work:

stop the wrong people early
let the right people through fast

3) Tune the chain: small lifts multiply

Once the funnel shape is right, the work is mechanical:

loosen gates that are accidentally tight
tighten gates that leak noise
remove ambiguity that causes drop‑off

The Compounding Maths

Similar to what we saw with the hyper growth framework, we can optimise the output of each gate in a funnel in a linear way (through trial and error), but optimising multiple gates at the same time could lead to exponential compounding growth in output from the system.

Baseline:

100 people enter
3 gates in the funnel
each gate passes 10% and rejects 90%
each conversion is worth $100

“With 3 gates at 10% each, only 0.1% make it through: 100 × 0.1 × 0.1 × 0.1 = 0.1 conversions.”

The Funnel Map: Volume × Sensitivity

So, how can you use AI to turbocharge your funnel’s output? And which tasks should be run by humans vs which should be run by AI?

Consider two axes:

Volume: how much volume your funnel is designed to handle
Sensitivity: how expensive it is to be wrong (money, trust, regulation, reputation)

This uses the same “stakes” axis as The AI-Jobs Framework.

High volume – Low sensitivity

Examples: ecommerce checkout, content subscribe, freemium signup

Who runs the gate: AI runs the gate. Humans supervise.

Do differently: Remove friction, automate variants, and run fast tests to maximise throughput.

High volume – High sensitivity

Examples: hiring at scale (company POV), compliance onboarding

Who runs the gate: AI runs the gate under human-defined rules.

Do differently: Route intelligently, keep decisions explainable, and enforce clear escalation paths.

Low volume – Low sensitivity

Examples: simple support, community join, lightweight onboarding

Who runs the gate: AI runs end-to-end. Humans handle exceptions.

Do differently: Set hard escalation boundaries and monitor re-contact or failure loops.

Low volume – High sensitivity

Examples: fundraising, partnerships, enterprise sales, senior hiring

Who runs the gate: Humans run the gate. AI assists.

Do differently: Operate account-by-account. Humans own trust, narrative, and final calls.

The Operator Layer

Use AI like a junior operator you can delegate to.

Give it a wrapper: allowed actions, forbidden actions, escalation triggers.

Four jobs:

Watch: track what turns into hires/purchases, and find the gates bleeding conversions.
Decide: score fit/intent, route, propose which gate to delete/amend next.
Write: draft what moves someone to the next gate (CVs, emails, landing copy, content).
Iterate: propose variants, track winners/losers, segment (don’t average).

Make It Profitable. Then Scale It.

Rule: tune until it makes more money than it costs to operate.

At a minimum:

value per conversion > cost per conversion
payback fast enough to reinvest

Once unit economics are positive, you have a self-funding loop. A perpetual motion machine:

reinvest profit into more high‑intent volume
monitor drift and fix it
compound into scale

If you want the extreme version, this starts to look like a swarm. See The Swarming Framework.

Organize your business around the funnel

Different teams must own specific parts of the funnel. Structure the organisation so each gate has clear ownership and KPIs, and ensure these are tracked and incentivised.

Where This Framework Applies

Hiring (candidate POV): apply → screens → interviews → offer
Hiring (company POV): sourcing → inbound → loop → close → onboarding
Ecommerce: impressions → clicks → PDP → cart → purchase → repeat
Freemium SaaS: visit → signup → activation → habit → upgrade
Content: impressions → clicks → subscribers → readers → referrals
Sales: leads → qualify → demo → proposal → close → renew
Fundraising: list → intro → meeting → partner → diligence → term sheet → close
Partnerships: outreach → discovery → pilot → contract → expand
Customer support: ticket → triage → diagnose → resolve → follow‑up

…and many more

In Summary

A funnel is a system. Its output is determined by the volume you feed in, the number of gates, and the pass rate of each gate.
Teams try to improve funnels by adding activity – more traffic, more steps, more tools. That usually increases entropy, not output.
The leverage comes from design. Name your gates. Remove the ones that do no real work. Fix the single gate that leaks the most before scaling volume.
AI doesn’t change this logic. It amplifies it. Used well, AI runs repeatable, high-volume gates and exposes where friction and drop-off occur. Used poorly, it scales noise and hides structural problems.
Small improvements across multiple gates compound. That is where exponential gains come from.
The job of leadership is to design the organisation around the funnel, assign clear ownership to each gate, and align incentives to the few metrics that actually drive funnel throughput and growth – while deciding which gates machines run and which require human judgment.

Check out some of my other Frameworks on the Fast Frameworks Substack:

Entity AI, swarms and the future of work (Asymmetric Podcast)

Fast Frameworks Podcast: Entity AI-Episode 8: Meaning, Mortality, and Machine Faith

Fast Frameworks Podcast: Entity AI - Episode 7: Living Inside the System

Fast Frameworks Podcast: Entity AI - Episode 6: Heard, Held, and Hacked: How AI Is Rewriting Intimacy

Fast Frameworks Podcast: Entity AI – Episode 5: The Self in the Age of Entity AI

Fast Frameworks Podcast: Entity AI – Episode 4: Risks, Rules & Revolutions

Fast Frameworks Podcast: Entity AI – Episode 3: The Builders and Their Blueprints

Fast Frameworks Podcast: Entity AI – Episode 2: The World of Entities

Fast Frameworks Podcast: Entity AI – Episode 1: The Age of Voices Has Begun

The AI implementation pyramid framework for business

A New Year Wish: eBook with consolidated Frameworks for Fulfilment

AI Giveaways Series Part 4: Meet Your AI Lawyer. Draft a contract in under a minute.

AI Giveaways Series Part 3: Create Sophisticated Presentations in Under 2 Minutes

AI Giveaways Series Part 2: Create Compelling Visuals from Text in 30 Seconds

AI Giveaways Series Part 1: Build a Website for Free in 90 Seconds

The delayed gratification framework for intelligent investing

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 2-5

The Fast Frameworks eBook+ Podcast: High-Impact Negotiation Frameworks Part 1

High-Impact Negotiation Frameworks: 5/5 - pressure and unethical tactics

High-impact negotiation frameworks 4/5 - end-stage tactics

High-impact negotiation frameworks 3/5 - middle-stage tactics

High-impact negotiation frameworks 2/5 - early-stage tactics

High-impact negotiation frameworks 1/5 - Negotiating principles

Milestone 53 - reflections on completing 66% of the journey

Spatial computing - a game changer with the Vision Pro