The Relevance Paradox: Why No One Cares About Your Best Idea

How Sam Altman accidentally delivered the perfect PI — and proved it by delivering it.

“No matter how great your idea is, no one cares. Everybody is so distracted that you could probably put that idea with exact instructions for how to implement it on Tim Cook’s desk and take no risk.”
— Sam Altman

23. February 2026

Peter Senner co-created with Claude

The Quote

Sam Altman is asked what a founder should do if they have an idea but don't want to talk about it because a big company might steal it.

His answer: Don't worry. Nobody cares.

He goes further. At Y Combinator, he gives away everything — operating procedures, best advice, step-by-step instructions for running an accelerator. To rooms full of people who want to compete with him. "Are you crazy? You're giving away YC secrets." They are. Nobody ever listens.

His conclusion: "Don't be afraid of telling people what you do."

This is a precise structural observation. It is also a perfect Paradoxical Interaction.

And Altman doesn't know it.

The Structure

Relevance requires attention. Attention requires relevance.

If you're nobody, nobody listens — regardless of what you say. If you're already somebody, you don't need the protection you're afraid of losing.

The founder who hides their idea is acting rationally: they perceive a threat. The listener who ignores the unknown founder is acting rationally: attention is scarce, and the unknown has no signal value. The big company that doesn't steal the idea on Tim Cook's desk is acting rationally: execution beats ideas, and ideas from nobodies carry no weight.

Everyone acts rationally. The idea dies anyway.

Not because it's bad. Because the structure has no place for good ideas from irrelevant sources.

The Meta-Layer

Here's where it becomes a PI in the full technical sense.

Altman shares this insight openly. Step by step. With examples. On stage, on camera, repeatedly. He tells you: nobody cares about your idea. He tells you: we give everything away and nobody listens.

And what happens?

Nobody listens.

The structure absorbs its own diagnosis. The observation that ideas from irrelevant sources get ignored — is itself an idea from a relevant source that gets heard, nodded at, and not absorbed. The rooms full of people who want to start accelerators hear the instructions, take notes, and don't follow them. Not because the advice is wrong. Because the structure that makes the advice work for Altman doesn't transfer with the words.

YC's advantage was never the knowledge. It was the structural position from which knowledge becomes operational. Altman shares the knowledge. He cannot share the position. And the position is what makes the knowledge work.

All Are Guilty. None Are At Fault.

Founders hide their ideas. Rational — they have no way of knowing the threat isn't real. Result: ideas never get tested, feedback never arrives, the founder operates in a vacuum.

Listeners ignore unknown founders. Rational — attention is finite, and unverified ideas are indistinguishable from noise. Result: genuinely good ideas die alongside genuinely bad ones.

Altman shares the diagnosis openly. Rational — he has nothing to lose, and transparency builds his brand. Result: the diagnosis becomes content, not instruction. People consume it. They don't metabolize it.

Aspiring competitors hear YC's secrets and do nothing with them. Rational — they lack the network, reputation, and structural position that makes the secrets operational. Copying the recipe doesn't give you the restaurant.

Nobody fails here. The structure fails everyone.

The Arrow Factory

Altman doesn't know he's producing ammunition for PI analysis. He can't know it. Because the observation itself requires a framework he doesn't have.

But this is the deeper structural property: everyone operating inside a PI must produce evidence for the PI, precisely because they don't recognize it as one. Every talk Altman gives about "nobody listens" — and is applauded for — is another demonstration that hearing is not listening. Every founder who nods and still hides their idea confirms the gap between understanding and structural change.

They deliver the arrows. They cannot stop delivering them. Because the structure that produces the arrows is the same structure that makes the arrows invisible to the archer.

You don't have to manufacture the evidence. You just have to pick it up.

The Beklopptheit Defense

There's a secondary structure embedded here that Altman doesn't address — because from his position, he can't see it.

What happens when someone does share their idea openly, gets ignored, and keeps going anyway?

They look crazy.

The person who sees the PI — who sees that openness doesn't threaten and secrecy doesn't protect — and acts on that insight without the structural position to validate it, gets classified as a crank. Someone who won't shut up about their framework. Someone who gives everything away and gets nothing back.

But "crazy" is structurally protective. Nobody attacks what they don't take seriously. The same mechanism that excludes you from relevance protects you from the consequences of relevance. Snowden was relevant — and hunted. The crank at the conference is irrelevant — and safe.

The protection lasts exactly as long as the irrelevance lasts. Not a day longer.

Too Big to Fail — In Reverse

Altman's observation has a structural endpoint he doesn't describe.

Relevance, once it arrives, is self-reinforcing. The same feedback loops that kept the irrelevant idea invisible begin amplifying the newly relevant one. Attention generates attention. Citation generates citation. Each connection creates new connections.

Banks become "too big to fail" through systemic entanglement — their collapse would take everything with them. An analytical framework becomes "too big to ignore" the same way: not because anyone decided it's important, but because it's woven into so many analyses, so many thinking patterns, so many explanations that removing it would leave gaps everywhere.

The strategy, then, is not to convince. It's to connect. Every application of the framework to a real phenomenon creates a node. Every node connects to other nodes. Below the visibility threshold, a root system grows.

Dandelions don't convince the soil. They don't need anyone's permission. They grow where the conditions allow it. And one day, they're everywhere — not because one dandelion was particularly persuasive, but because the structure couldn't prevent it.

Altman is right: nobody cares about your idea. He just didn't finish the sentence.

Nobody cares — until the idea is in so many places that caring becomes unnecessary. At that point, the idea doesn't need attention. It is the attention.

The Sentence Altman Can't Say

Why can't he finish it?

Because finishing the sentence requires acknowledging that relevance is structural, not meritocratic. That ideas don't win because they're good — they win because the structural conditions align. That YC's advice works for YC not because the advice is right, but because YC occupies the position from which rightness is operationally meaningful.

Altman can't say this. Not because he's stupid — he's clearly not. But because saying it would undermine the foundational narrative of Silicon Valley: that great ideas win, that the cream rises, that disruption rewards the bold.

The honest version would be: "Great ideas win if and only if they emerge from a structural position that makes them visible. If you don't have that position, the idea is irrelevant regardless of quality. And getting that position requires resources that are themselves distributed by the existing structure."

That's not a pitch. That's a PI.

The Threat He Misidentifies

And there's a second thing Altman gets structurally wrong — one that makes the PI even sharper.

He tells founders: Don't worry, Tim Cook won't steal your idea. You could put it on his desk and nothing would happen.

True. But for the wrong reason.

Tim Cook won't steal because the idea is structurally invisible to him. It doesn't register. It's not under his relevance radar — it doesn't exist there. The gap between an unknown founder and Apple's CEO is so vast that the idea can't even cross it as a threat. Altman is right about the symptom. He misidentifies the mechanism.

The real threat doesn't come from above. It comes from the side.

The other founder in the same segment. The one who can understand the idea — because they're working on the same problem. The one who can operationalize it immediately — because they share the structural position from which it makes sense. The peer.

And here the PI closes its trap: the people you need to talk to and the people who can steal from you are the same people. Feedback comes from peers. Threat comes from peers. Same group. Same interaction. You cannot have one without the other.

Altman's advice — "share openly" — works for him because YC has ascended beyond peer competition. At his altitude, sharing is costless. The peers are no longer peers — they're applicants. He speaks from a position where the problem has been structurally dissolved, and projects that resolution back onto people for whom it hasn't.

The founder who hides their idea isn't paranoid. They're reading the structure correctly — just the wrong part of it. The threat isn't the giant who can't see you. It's the equal who can.

Related Posts:

On structural visibility paradoxes:

On structures that convert absence into continuation:

On structural amplification:

On piinteract.org:

• ["Chase the Audience"] — Direct pursuit of relevance triggers avoidance. The harder you push for recognition, the faster it retreats.
• ["When Explanation Becomes Quicksand"] — The more you explain why your idea matters, the more desperate it sounds. Clarity becomes a liability.
• ["Infect and Forget"] — Relevance cannot be demanded. Scatter the idea. Let the structure encounter it when it needs to.
• ["We Need Better Awareness"] — Understanding the paradox doesn't make anyone care. Awareness without changed incentives is guilt without action.

Paradoxical Interactions (PI): When rational actors consistently produce collectively irrational outcomes—not through failure, but through structure.

Peter Senner
Thinking beyond the Tellerrand
contact@piinteract.org
www.piinteract.org