Editor's Note: AI is making the differences between startups increasingly blurry. Application companies are starting to build infrastructure, infrastructure companies are moving into workflows, and all companies are repackaging themselves with new concepts. Products, interfaces, technological narratives, and fundraising pitches are becoming easier and easier to copy.
What this article truly wants to discuss is: when products are no longer scarce, what becomes a company's moat? The author's answer is the organization itself.
The best companies don't just hire talented people; they create a structure within which a certain type of person can truly flourish. What makes OpenAI, Anthropic, and Palantir unique isn't just their technology or market position, but the fact that they have built new organizational forms around new ways of working, thereby shaping a new type of talent.
The inspiration for founders is this: don't just ask, "How do we tell a better story?" Instead ask, "What kind of person can only truly become themselves here?" A truly attractive company doesn't just offer high salaries; it offers talent a clear mission, authority, growth path, and tangible rewards.
For individuals, this article also reminds us that "being chosen" and "being seen" are not the same. The former is an emotion; the latter is structure. A company truly worth investing your time in isn't just one that makes you feel special, but one willing to codify your value into roles, resources, decision-making power, and long-term gains.
AI will make many things easier to copy, but it won't make building a great company any easier. The real moat of the future may no longer be just technology or product, but whether a company can create a new organizational form that allows the right people in the right positions to continuously generate compound returns.
The following is the original text:
It's clear that everything in AI is converging. Companies I never imagined would compete are now sitting at the same table. The application layer is collapsing into infrastructure, and infrastructure companies are also moving up into workflows; almost every startup is repackaging itself as some kind of "transformation company."
These terms change every few months: context graphs, action systems, organizational world models. A new category is named, and all company websites quickly absorb it; within weeks, the market is flooded with companies claiming to be the "inevitable platform for the future of work."
When models advance rapidly, interfaces converge, and the speed of product iteration becomes cheap, the visible parts of company building become increasingly easy to imitate. What's truly hard to replicate are the underlying organizational mechanisms: how a company attracts exceptional people, organizes their ambitions, focuses judgment, distributes power, and turns work into a compounding system others cannot copy.
The best companies have always known that people are not just an input to the company; people *are* the company. But in the AI era, this becomes more acute because everything else changes so fast. If products can be copied, categories can be renamed, and technical advantages can collapse in months, then the truly long-term question becomes: what kind of organization have you built around those capable of creating the company?
The organizational form itself is becoming the moat.
Great Companies Are Essentially Organizational Inventions
The most important companies are, in fact, organizational inventions. They create a new type of institution around a new way of working; and in that process, they enable a new type of person to emerge.
OpenAI is neither an academy, nor a corporate research lab, nor a traditional software company. Its core organizational activity is frontier model training. Safety, policy, product, infrastructure, and deployment all revolve around this gravitational center. This structure changes what kind of person a researcher can become within it: someone working at the edge of science, product, geopolitics, and civilizational risk simultaneously.
Palantir, meanwhile, invented a new operational institution for broken systems. "Frontline deployment" isn't just a sales motion; it's simultaneously a status hierarchy, a talent model, and a worldview. It placed work that elsewhere might be seen as low-status—sitting with clients, absorbing organizational chaos, translating political realities into product—at the very center of the company. It created a protagonist: someone who cannot be easily categorized as a software engineer, a consultant, or a policy expert, but can move between all three.
These companies cannot be fit into existing boxes. The people who create them cannot either. Great companies aren't just places where talented people gather; they are structures that allow a certain kind of talent to finally express itself.
Organizational Form Determines Who Can Exist There
The world's best companies don't just compete on category, market, or compensation. They compete on identity.
Ambitious people typically care intensely about a few things: feeling special, proximity to power, becoming undeniable, preserving optionality, belonging to a mission, being in the room where history is bending. But they often don't yet know which of these goals they are truly optimizing for.
This is why the strongest institutions identify talent very early, recruiting from top universities starting as early as freshman year. They reach people before their self-concept is fully formed—before they know what they want to be famous for, what they truly believe, or how to separate "work they are good at" from "the person they want to become."
A great company gives these ambitions a language. It says: You've been circling around something without knowing how to name it; and that thing can happen here. You can become the person who moves the Mars timeline forward, who is in the room when the frontier pivots, who can operate inside broken institutions, whose work eventually becomes undeniable.
This is why great institutions are, at their core, containers built around a certain type of person.
Many companies compete on cash. But for legendary companies, cash is one of the least interesting forms of talent competition—perhaps with exceptions like Jane Street or Citadel. Cash can get someone to sign, but it rarely truly converts a person. The best people are often most loyal when a company offers something more specific than money: a path to the version of themselves they already wanted to become, or haven't yet realized they want to become.
Every emotional commitment must also have a corresponding structural commitment. If a company says close customer contact is important, but customer-facing work is low-status internally, that commitment is fake. If a company says ownership is important, but decision-making power is highly centralized, that commitment is fake. If a company says mission is important, but that mission offends no one, screens out no one, and requires no sacrifice, that commitment is also fake.
So, what do people really want to feel?
People want to feel special: scarce, seen, irreplaceable. This recruiting narrative lands as: Only you can do this, only you are unique enough, should come here and build it together. It targets the quiet insecurity in many high-performers' hearts: they doubt their excellence is fragile, suspect others could do the job, doubt they've been truly seen. This promise only works in an organization small enough that one person can actually change its trajectory.
People want to feel chosen by fate: that their life is bending toward something inevitable. Anthropic is the cleanest current example. It conveys: We are one of two or three companies that will decide how this technology is deployed safely, and the people in this room are the ones doing it. This emotion is only credible if a company is structurally, indeed, in that position.
People want to feel they haven't missed out: that they are in the room where compounding is happening. Look at how many CTOs from iconic companies Anthropic has recruited just this quarter. Talent density itself is a choice of organizational form: it depends on how a company hires, pays, organizes work, and concentrates the best people in the same physical space.
People also want to feel they still have something to prove. Like the investment banker, polished, certified, told they're excellent, but starting to suspect none of it really proves anything. Or, people want to preserve optionality. McKinsey took this to an extreme: generalist project allocation, two-year analyst cycles, and the optionality to explore different industries—after all, who knows what a 21-year-old really wants to do?
Of course, people also want proximity to power and status.
Some people want sacrifice, wanting that sacrifice to point to something bigger than a salary. Most companies in the past called this mission, but the way it truly works is more like a quasi-religious community formed around deep team belief. The value proposition of some new-generation lab-like companies is sharper than the last wave's mission statements because they explicitly take sides. Open source means you stand against closed labs; sovereign AI means you oppose the assumption that one country's models run the whole world. The strongest missions often make some people refuse to work there—precisely because that means it will also make the right people desperately want to join.
People are people, after all. The best companies usually pick one, at most two, specific emotions their ideal candidate is most hungry for, and have already built the corresponding organizational form for that person.
The Real Question Founders Should Ask
For founders, the real question isn't: How do we tell a better story? It's: Which kind of person can only truly become themselves here?
Most companies tell the literal version of what they do. We're building a model. We're building a rocket. We're making a CRM for X. We're automating Y. These statements may be accurate or honest, but today, accuracy is no longer enough to recruit exceptional people.
The best companies today express themselves from a higher altitude. They describe the change their existence enables: an industry will be revived, an institution rebuilt, a civilization-level bet won, a certain type of human endeavor made possible for the first time.
Sometimes, people mistake this "extra altitude" as just marketing, and it's different from a fundraising narrative. But your story's posture must match your company's organizational form. A grand story inside a small organizational form sounds like empty words; a small story inside a grand organizational form will make the best people pass you by. What candidates are really assessing is the alignment between the two, even if they can't articulate it clearly.
If you believe close customer contact is a moat, then customer-facing work must have high status.
If you believe speed is a moat, then decision-making power must be pushed to the edges.
If you believe talent density is a moat, then average people cannot be allowed to set the company's operating pace.
If you believe deployment capability is a moat, then those closest to reality need to have power, not just responsibility.
For Those Making a Choice
For those choosing where to direct the next chapter of their life, the lesson is different. You're entrusting years of your life to a specific person's vision and a specific organizational form, and the hiring process is very bad at revealing either.
Hiring will show you the narrative, the mission, the talent density, and the imagined future. It rarely shows the real power structure, and almost never shows how people will act under pressure.
These things usually only appear later: when the company is under stress, when your work becomes inconvenient, when you ask for something the company didn't originally want to give you, when their belief in your potential must translate into title, power, economic interest, scope, or resources.
For ambitious people, emotional validation can make you feel like an owner long before you actually get ownership. High-performers might work like founders, absorb uncertainty like executives, internalize the mission like principals, yet receive the pay and power of an employee. The company captures founder-level intensity, and the individual gets a sense of belonging. When structure eventually catches up, this exchange can be beautiful; but when structure doesn't, it becomes asymmetric.
Older people will warn you: You're paying in identity what should be paid in structure. You substitute "specialness" for title, "proximity to power" for actual power, "reassurance" for economic interest, "trust me" for written mechanisms. This is why one can feel deeply valued yet materially and structurally stuck.
Of course, employees can be rewarded through many levers, like ownership and compensation. But the most dangerous promises are often priced in time. This will get bigger over time. You'll have more over time. The structure will catch up over time.
But time never gives notice when it leaves. You arrive at some later version of your life and realize the promise made in the future tense never materialized—or maybe it did.
For ambitious people, you must realize that "being chosen" and "being seen" are different. Being chosen is emotional: You're special, we believe in you, you belong here. Being seen is structural: This is your scope, this is your power, this is your economic participation, this is your decision-making authority, this is what changes if you succeed.
If you truly have potential, go where someone is willing to truly see it. Go where they are willing to codify your value into the organizational structure itself.
The New Moat
You can, of course, read all of this cynically. You can think every recruiting narrative is manipulation, every mission a cloak, every company trying to make you feel special to rent your life at a discount.
But our psychology needs to believe in something. We want our work to mean something, our sacrifices to point to something, our talents seen by those who can truly use them. This doesn't make us naive; it makes us human.
Great companies have always been new containers for this need. They aren't just vehicles for products or profits; they are structures for ambition.
Silicon Valley loves its categories: technical, non-technical, researcher, operator, founder, investor, missionary, mercenary... and then forgets that most truly great people don't live in one box. They live between boxes, borrowing from one, breaking another, combining things that shouldn't touch, eventually creating a new form that others later mistake for obvious.
The opportunity now is not to become the next OpenAI, Anthropic, Google, Palantir, or Tesla. It's to ask: What kind of company couldn't have existed before? And what kind of person has been waiting for that company to appear?
AI will make many things easier to copy: product interfaces, workflows, prototypes, fundraising pitches, even early speed. But no matter how many narratives claim AI makes building institutions easier, it won't make creating a *new* institution easy. It won't let you easily build the kind of organizational form that can gather the right people, give them the right power, bring them close to the right problems, and let their judgment compound over time.
The old talent market rewarded companies that made people feel "chosen." The next-generation talent market will reward companies with organizational forms the old market couldn't produce. And the people within them will become people the old organizational forms couldn't have shaped.
