The OpenAI Pentagon Deal Isn't a Mistake. It's the Seneca Trap. And Sam Altman Fell For It.
Sam Altman told us exactly what he thought of this deal before he made it.
"Opportunistic and sloppy." His words. Not a critic's framing, not a hostile journalist's read, not a disgruntled employee's leak. Altman, describing exactly this kind of military entanglement, called it by name before the contract existed. He had the right diagnosis. He had the right vocabulary. He had the credibility of someone who'd spent years insisting OpenAI is different, that the mission is real, that AGI in the wrong hands is the actual thing we're supposed to be afraid of.
And then he signed it.
Seneca, writing to his friend Lucilius in the last years of his life, made an observation that's been sitting in Latin for two thousand years waiting for this moment. The substance: we do not stumble into wrong. We reason our way there. The more capable the mind, the more elaborate the path to the thing you already knew you shouldn't do.
He was watching the Roman Senate. Men who could quote Stoic philosophy at dinner parties and vote for atrocities before dessert. Men who knew the vocabulary of virtue perfectly, deployed it perfectly, and then made the corrupt call, slept fine, and made another one tomorrow. His diagnosis wasn't weakness or obvious corruption. It was the gap.
The gap between knowing and doing. The hours, days, or board meetings between "this is wrong" and "here's why it's actually right." The space where the mind goes to work not on the question — that's settled — but on the justification for the answer you've already decided on.
The most dangerous thing about intelligent people, Seneca saw, is their ability to construct good arguments for bad conclusions. Stupidity stumbles. Intelligence walks through the gap with its head up, explaining exactly how this case is different from the thing it said it would never do.
The thing about the Seneca trap is it doesn't feel like a trap from the inside. It feels like maturity.
OpenAI needs compute. To get the compute, the data center deals, the chip access, the regulatory relationships that keep you ahead of model labs building the same capabilities without any safety protocols, you need the Pentagon. You need to be in the room. "Opportunistic and sloppy" is easy when the stakes are abstract. When it's concrete, when the alternative is other organizations building the same weapons-adjacent AI without your guardrails, it gets called something else. Survival. Realism. Understanding that you can do more good inside the tent than shouting principle at a closed door.
This is not cynicism. That's the thing Seneca saw and that we keep missing.
Altman isn't cynical. He built this thing. He has watched OpenAI go from a scrappy nonprofit with actual mission urgency to the most consequential AI company on earth, and I'd bet he genuinely believes OpenAI inside the tent is better than the alternative. He believes in the mission. He believes in being in the room.
The trap isn't what cynics walk into. It's what believers walk into when the institution becomes the self.
The moment your organization becomes your identity, really becomes it, not as a metaphor but as the actual ground your sense of self stands on, something shifts in how you reason about threats to it. Every threat to OpenAI becomes a threat to the mission. Every compromise that keeps OpenAI alive and growing feels like it serves the thing it's actually betraying. The mind that named the trap is now using that naming as cover.
"I know what this looks like from the outside." That sentence isn't prevention. It's preparation. It's the first line of the justification.
This is the mechanism Seneca was pointing at. Not moral failure in the obvious sense. The merger of self and institution, and the reasoning that follows automatically once that merger is complete. The smarter you are, the cleaner the path through the gap. The more principled your public history, the more credible the exception sounds. Altman had years of on-record concern about exactly this kind of deal. That history became the collateral he drew on to make the compromise legible to himself.
You've done this.
Not with a Pentagon deal. With the thing you knew before you did it.
The hire you knew was wrong before the interview ended, but you needed the headcount and you told yourself she'd grow into the role. The client you knew wasn't a fit, but the retainer was real and you told yourself you'd fix culture with the cash. The product decision you knew was pulling you away from your actual thesis, but the market wanted it and you told yourself Q3 was when you'd get back to what you were building.
You knew. Somewhere before the decision was a decision, you knew. And then your mind went to work. And by the time you made the call, you had reasons. You could explain it to a board, a co-founder, yourself at 2am. The gap had been crossed and paved over and you were standing on the other side feeling like someone who thought it through.
Seneca isn't writing about people who don't care about doing right. He's writing about people who care deeply, who have sophisticated reasons for why this particular case is different, why the exception isn't an exception, why the thing they said they wouldn't do is actually in service of the thing they're trying to protect.
The believer is in the gap. Not the cynic.
The Seneca letter doesn't end hopefully. He doesn't say you can close the gap or build a rule that keeps you out of it. He says you can see it. And seeing it while you're standing in it is the only real advantage, because the gap works by being invisible.
Altman saw the gap years ago and named it publicly. What he didn't account for is that naming doesn't spring the trap. The naming can become part of the preparation. "I know what this looks like from the outside" is not the same as "I'm not going to do it." When the institution is the self, the naming is more often the first draft of the justification than the thing that stops you.
The question isn't whether you're someone who knows right from wrong. You are. The question is what happens in the space between knowing and acting, while the machinery runs.
Because it's running right now. On something. The version of you that's been building the case since before you knew there was a decision to make.
The gap is already open.
Jaxon Parrott