OpenAI risks White House relationship with hiring of ‘nuisance’ AI policy executive: sources

OpenAI just hired the one guy in AI policy circles practically guaranteed to irritate the people currently writing the rules. Sources say Trump administration officials were already annoyed with this person over a pitch to create "regulatory risk" for American firms as a way of scaring them off Chinese AI models. That's not a subtle ask. It's asking Washington to punish domestic companies now, on the theory that it might discourage them from doing business with Beijing later. Whatever the merits, it's the kind of idea that reliably gets you labeled a nuisance in a town that's trying to move fast on AI policy, not slow everyone down with new liability traps.

The bigger tell here is what it says about OpenAI's instincts. This is a company that has spent the last year cultivating access in Washington, courting this administration specifically, and positioning itself as the adult in the room on AI governance. Then it turns around and brings in someone who picked a fight with the very officials it needs to stay friendly with. Either nobody at OpenAI thought this through, or they did and decided the hire was worth the friction anyway.

Neither option is especially reassuring. If it's the former, it's sloppy management of a relationship that matters enormously to the company's future, especially with federal AI rules still being written. If it's the latter, it suggests OpenAI thinks it can push the White House toward a more restrictive, risk-laden regulatory posture on Chinese competitors from the inside, even after officials already made clear they don't want that fight framed that way.

We'd rather see American AI companies working with this administration to keep the US ahead of China through innovation and smart export controls, not lobbying for new domestic red tape dressed up as toughness on Beijing. There's a real China challenge in AI. Manufacturing "regulatory risk" against your own industry isn't how you win it.