The Board Blinked
OpenAI fired its CEO to protect humanity and humanity's employees said no thanks.
The board blinked. Altman came back. The theory about his APEC comments triggering the firing was never confirmed but the broader point — that the board panicked about acceleration and then lost — became the definitive narrative. The floor shifted and the people who tried to stop it got removed.
Two days before the board fired him, Sam Altman stood at APEC and said, essentially, that we are close — that he's felt the floor of AI shift under his feet four times in OpenAI's history and we are approaching one of those moments again.
The current favorite theory is that this statement is why he's no longer CEO.
Not the statement itself, necessarily. But what was behind it — some internal breakthrough serious enough that someone, apparently, felt the need to tell the board. The board, in their infinite wisdom, decided the correct response was to fire the person most capable of navigating whatever they'd just been told about.
This is the kind of decision you make when you have a mandate to benefit all of humanity and no particular plan for how to do that.
What happened next is almost too stupid to type. The employees — the ones who would actually build the thing the board is apparently terrified of — sent a letter saying: reinstate Sam and Greg or we all quit. Over seven hundred signatures. Not a handful of loyalists. Basically the whole company.
The board, which fired a CEO to save the world, is now apparently preparing to resign and give the CEO his job back, because otherwise the world-saving organization ceases to function.
There is a version of this where the board was right. Where something real was found, something that warranted the nuclear option. And then seven hundred people said "we don't care, we want our guy back" — and the board, the only adults in the room by their own accounting, folded.
The governance structure designed to prevent an AI catastrophe lasted about 36 hours against a strongly-worded letter.
I don't know what Q* is. Nobody outside OpenAI knows. Maybe it's nothing — a research artifact, a capability that sounds more alarming in a memo than it is in practice. Maybe it's something. But I do know that whatever the board thought they were doing, the outcome is: Sam Altman is probably going to walk back in on Monday, the people who threatened to follow him to Microsoft will stay, and the board that tried to stop it will be gone.
The control problem, as it turns out, was the board.
Counterpoints
Push back, extend the argument, or sharpen it. New counterpoints go through review before they show up here.
No approved counterpoints yet.