The COROS Whitepaper · Volume Two
The Buffalo Has Disappeared
The core operating system for the era after machines.
Published October 27, 2026 · Seattle, Washington
How this paper works
Volume one named the cost. Volume two says what the operating system is.
Volume one, Can AI Save Humanity? (August 2025), diagnosed the collapse of workability and named coordination waste. Near its end it promised a group AI for teams and companies. Volume two is where that promise comes due. It carries forward what volume one got right, and it says what the operating system is. Thinking recursed.
The title is shared with Saqib’s book, deliberately. The paper is the company’s argument. The book is Saqib’s story. Same buffalo, two doors, each with its own subtitle.
Companies come first in this paper, then the individual. What you are reading is the founding document: the thesis, the prologue and the first chapter. The chapters that follow are outlined at the end, and they will appear here as they are finished.
The thesis
Six claims, in order.
-
For a century we paid people to act like machines. AI just ended that deal.
The deal built the modern world, and it hid a cost. Machines do not have moods, so we never managed the moods. Machines do not make promises, so we never watched the promises. We built systems that record what happened, and a company breaks in what was never recorded: a request nobody made, a promise nobody confirmed, a disagreement nobody voiced, a decision made in the room and undone in the corridor. The waste in that layer is the largest in any company and the only one that appears on no report. Volume one named it: coordination waste, carried in moods, living in language.
-
It could not be seen until now, because it lives in language, and until now nobody could be present in all of it.
Two things happened at once. AI ended the old deal; machines are now better at being machines than we ever were. And everything that matters in a company began to be transcribed, in one tool or another. For the first time, help can be present in every conversation and remember every promise. Most companies will answer this moment with fewer people. That is the worst use of the most powerful technology of our lifetime. The companies that win will use it to make their people more capable than they have ever been.
-
Seeing action in language takes a lens, not a bigger model.
General AI is built to read how people feel and what they ask: sentiment, intent, how someone seems. Asked about a meeting, it returns a summary, and a summary of a meeting is a butchering of the meeting. What decides whether a company moves is something else: what was promised, by whom, by when; which request was never made; which commitment is slipping; which disagreement was swallowed; and what mood the room was in when it was said. They read sentiment. We read commitments. Our lens comes from forty years of research on language and human action, carried by our predecessors at Berkeley and Stanford, and from more than ten thousand hours of our own live practice. We have trained an AI to see with it.
-
COROS is the core operating system for how people work together.
A master coach in every conversation of your company, and your own master coach in your pocket. It reads the company’s action state, the living record of every loop of action, and above all three things: moods, breakdowns and completions. It serves each person privately, through two memories: theirs, which holds their conversations and is theirs alone; and the company’s, which holds only what is shared between people, never a personal detail. It carries the company’s soul, what the company has declared it is up to, into every conversation. And it runs under a constitution no contract can change. Integrity as a service. People tell COROS the truth only because it is safe to, and the truth is the only thing here worth having.
-
We have tested the reactive AI, and the results are in.
The lens produced a $68 million year at a half-billion-dollar franchise before a line of code was written. Measured independently across hundreds of participants, 82 percent took bold action within days. The product is live and paying, and it counts its own work in the open. Now we are building the mobilization: the first conversation, follow-through, the meeting companion, commitments, the company picture, and agents fitted to each company’s own practices, with two design partners in live companies.
-
The soul of a company is each person’s full success in their job, and their full satisfaction.
That is what COROS serves, one person at a time. It never works on anyone’s behalf with other people. It is a polished mirror in which each person sees their own part, and when one person changes, the room changes with them. That is why it begins with the individual, why individuals become the champions who open companies, and why, when a person leaves a company, their mentor leaves with them.
Prologue
After this nothing happened.
In the last years of his life, an old man sat down with a writer and told him the story of everything that had happened to him. His name was Plenty Coups. He had been the last great chief of the Crow nation. As a boy he had fasted alone on a mountain and been given a vision. As a young man he had counted coup: ridden at an enemy, touched him with a slender wooden stick, and lived to tell it. As a chief he had carried his people through wars, treaties, the railroad, and the hard arithmetic of survival on a shrinking land. He told the writer all of it, in order, for days.
Then he came to the year the buffalo were gone, and he stopped.
The writer asked about the years that followed. The chief would not speak of them. He had lived more than forty years past the buffalo. In those years there had been births and deaths and seasons, there had been a treaty and a war across the ocean, there had been grandchildren. But nothing, he said, had happened.
The philosopher Jonathan Lear spent a whole book on that sentence, and his reading is the one that stopped us. The Crow had not lost a food supply. They had lost the world in which their acts meant anything. Courage, for a Crow man, had a shape: you drove your coup stick into the ground and you did not move from that spot. You had declared that ground. You held it or you died on it. When the buffalo went, there was no longer any ground of that kind anywhere. A man could still perform every motion of courage, perfectly, exactly as he had been taught, and it would mean nothing. Not because he did it badly. Because the world that gave it meaning was gone.
He committed his people to a flourishing future he could not yet describe, and he held.
What Plenty Coups did next, Lear called radical hope. He accepted that the world he had been raised in was finished. And he committed his people to a flourishing future he could not yet describe, and he held, and the Crow found a way, and they are still there.
We open a paper about companies with this story because it is the most precise description we have found of the moment we are in.
Chapter 1
The deal that ended.
1911
In 1911 an engineer named Frederick Winslow Taylor published a short book about how to run a factory, and the modern world was built on it. The idea was simple and it worked. Break the work into parts. Measure each part. Fit each person to a part. Pay for the part done. Taylor was not cruel. He thought he was ending the arbitrary rule of foremen, and in some ways he was. But the book carried a picture of the human being inside it, and the picture was a machine part.
The book carried a picture of the human being inside it, and the picture was a machine part.
That picture became a deal, and the deal lasted a hundred years. You act like a part. Follow the process, hit the number, leave the rest of yourself at home. In exchange, we pay you, and we keep paying, and the paying is steady. The factory made the deal first. Then the office inherited it, and so did the hospital, the bank and the university. We built software to run it: contacts, tickets, deals, dates, performance reviews, objectives and key results. Every one of those systems recorded what a part had done. None of them had a place to put what a person had said.
For most of the century the deal held, and it was a good deal for most of the people who took it. It put a generation into houses. It let a boy from a small town become an engineer and a manager and a man with a pension. We are not writing to condemn it. We are writing because it is over.
Here is how it feels from the chair at the top, in the year it ends.
A chief executive walks into the Monday meeting. She has strong people around the table, people she hired herself. The strategy is sound; she has argued it with the board and won. The market is real; customers are asking for the thing. And the company does not move the way it should. A commitment made last week has slipped, and nobody says so. A decision was made in this room on Monday and quietly undone in the corridor by Wednesday. Two of her best people have stopped speaking to each other about the substance and now speak only through her. Her calendar has filled with meetings that exist only to reconcile people. And last month the best engineer she had, the one with the most options, left politely, for a reason that sounded like something else.
She cannot say any of this out loud. Not to the board, who would hear risk. Not to her directs, who are half the subject. Not to the friend who tells her it sounds like a great problem to have. So it comes home with her, and the evening is spent litigating a conversation nobody else in the room witnessed.
Nothing dramatic is happening to her company. That is the problem. It is working, and it is not compounding, and every system she owns says it is fine.
She has tried the familiar remedies. A reorganization, which moved the boxes while the same conversations produced the same results. A leadership program, which taught concepts to people who already knew them. A new tool, which recorded the work more accurately while the work kept falling apart. Each one created activity. Then the company settled back where it was.
The problem was never the process. It lives in the conversations the process runs on.
Every result her company produces began as a conversation: someone asked, someone promised, someone decided, someone delivered. When those conversations are clear, work flows and trust grows. When they are muddled, everything becomes friction, and no process on the surface makes up for it. For a century we treated people as machines, and machines do not have moods, so we never managed the moods. Machines do not make promises, so we never watched the promises. The deal hid a cost, and the cost is paid in action that does not move.
Then, in the span of about three years, the deal ended.
The machines got better at being machines than we ever were. The part of the job that was following the process and hitting the number, the part the deal was made for, is now done faster and cheaper by software that does not go home at night. Every chief executive knows it. Her board is asking what she is doing about AI. And nearly everyone offering her an answer is pointing it at the same place: headcount. Fewer people, more software, call it transformation.
We think that is the worst use of the most powerful technology of our lifetime. It saves a line on the budget, and it spends the trust of everyone who stays. And it misreads what the moment is. When a deal like Taylor’s ends, everything it was hiding comes out into the open. What is left, once the machine work goes to the machines, is everything no system ever recorded: how people navigate action, make commitments, deliver, and create value together. That layer was always where the company actually lived. It was simply invisible, because it lived in language, and nobody could be present in all of it.
The worst use of AI is fewer people. The best use is more capable ones.
Plenty Coups could not have kept the buffalo. Nobody could. His greatness was in what he did once they were gone: he did not keep planting the coup stick in ground that was no longer there, and he did not let his people fall to the ground with their hearts. He turned toward a future he could not yet describe, and he held.
The companies that win this decade will not be the ones that answered the end of the deal with fewer people. They will be the ones that answered it with people more capable than they have ever been: more awake, more able to ask and promise and repair, more able to move together.
This paper is about what the deal hid, why it could not be seen until now, what it takes to see it, and what a company does once it can.
What follows
The chapters to come.
The paper is written in the open. The prologue and the first chapter are above. The rest appears here as it is finished.
- · Above
Prologue. After this nothing happened.
Plenty Coups, the coup stick, and Jonathan Lear’s reading: the Crow did not lose a food supply, they lost the world in which their acts meant anything. Radical hope: committing to a flourishing future one cannot yet describe, and holding.
- 01 Above
The deal that ended.
Taylor’s factory, the century-long bargain, and the office that inherited it. A CEO’s Monday: strong people, sound strategy, real market, and the company still does not move. AI ends the deal, and the worst answer to it.
- 02 Coming
What the deal hid.
Coordination waste, in volume one’s own words. The evidence of the cost, read as one problem rather than four. The historical roots: Taylor, the Cartesian inheritance, Toulmin’s Cosmopolis.
- 03 Coming
Why the obvious fixes return to baseline.
Reorganizations move the boxes while the same conversations produce the same results; training teaches concepts, not practices; tools record what happened; coaching floats in mindset. And the new obvious fix: AI pointed at headcount.
- 04 Coming
Why it could not be seen before AI, and why general AI still cannot.
Everything that matters is already in people’s language, and language is now transcribed. Generative AI as the cure emerging from the disease. Then the turn: they read sentiment, we read commitments.
- 05 Coming
The lens: action can be observed in language.
Austin’s discovery that some words do not describe the world but change it. A company as a network of requests, promises, assessments and declarations, carried in moods. The loop of action and where work gets stuck in it. The Coordinator of the 1980s, and its lesson: trust is built into the design, not promised in the brochure.
- 06 Coming
The core operating system.
COROS for Companies. A master coach in every conversation. The company’s action state, defined: moods, breakdowns, completions; not a task board. Two memories. Not a second product: your own COROS with your company’s shared memory switched on. The soul, defined, and given a voice. The constitution: Integrity as a service, with the five promises.
- 07 Coming
What we have tested, and what comes next.
The reactive AI: Planet Fitness, the Capria portfolio, the Novi Research results, the live counts. Then the mobilization: the first conversation, follow-through, the meeting companion, commitments, the company picture, agents fitted to each business. Two design partners building to the public promise in live companies.
- 08 Coming
How a company starts.
An hour, and listening. The company’s soul written with its own people. One team, one real challenge, free to start. Human mentors by the session, certified by our partner Conceivian. The price principle: one percent of what your people earn, spent on their growth.
- 09 Coming
The individual: a master coach in your pocket.
Where it began, and where champions come from. The first conversation, memory, the mentor who knows what you are after. Coaches, mentors and consultants as the second channel. Your mentor leaves with you.
- 10 Coming
The invitation.
Investors, engineers, coaches, founders and operators, mobilizers. The bet: not on productivity, on the capital trapped in coordination waste, and on people who are more capable than they have ever been.
- · Coming
Letter from the CEO: what we learned since volume one.
The year between the papers: the operating system, the action state, the constitution, and the two companies.
Saqib Rasool
Co-Founder & CEO, COROS AI
Seattle, Washington · October 27, 2026
Notes & sources
- 1.Frank B. Linderman, Plenty-coups, Chief of the Crows (1930). The chief’s words on the buffalo are as Linderman recorded them.
- 2.Jonathan Lear, Radical Hope: Ethics in the Face of Cultural Devastation (Harvard University Press, 2006). The reading of Plenty Coups in the prologue is Lear’s.
- 3.Peter J. Denning, Turing’s Mistake, which carries the same story to a neighboring argument.
- 4.Frederick Winslow Taylor, The Principles of Scientific Management (1911).
- 5.Volume one’s notes and sources carry forward and stand behind the claims the thesis repeats from it: the Planet Fitness franchise, the Novi Research measurement, and the hours of live practice.