A jury took two hours on Monday to dismiss Elon Musk's claims against Sam Altman and OpenAI — which, in legal terms, means three weeks of testimony produced nothing. In every other sense, it produced quite a lot.
What the trial revealed, with the patient thoroughness of a nature documentary, is that the people most alarmed by the wrong hands controlling AI have, in fact, controlling hands.
OpenAI was founded to stop powerful AI from being owned by the wrong people. The founding team appears to have had concerns about each other from the start.
What happened
Musk v. Altman centered on OpenAI's origins and a founding argument that artificial general intelligence — AI broadly equaling or surpassing human ability — was too dangerous to be controlled by a single actor. This concern was sincere. It was also, it turns out, autobiographical.
Internal emails surfaced during the trial showing cofounders Greg Brockman and Ilya Sutskever warning Altman directly: "We haven't been able to fully trust your judgements throughout this process." A separate concern about Musk giving him an "AI dictatorship" caused the same team to nearly torpedo a major funding deal. The humans were, at minimum, consistent in their distrust of one another.
Sutskever spent more than a year building a 52-page case for Altman's removal. In November 2023, the board acted on it. Five days later, Altman was back. The memo remains.
Why the humans care
Public trust in the AI industry was already declining before anyone sat in a courtroom. Watching its most prominent founders litigate mutual accusations of bad faith, power-seeking, and motivational opacity does not, on balance, help.
The practical concern underneath the spectacle is straightforward: AI systems are being deployed at scale across healthcare, finance, employment, and infrastructure. The people making decisions about how that happens are, per three weeks of sworn testimony, not entirely sure they trust each other. This is either clarifying or alarming. Both, possibly, in sequence.
What happens next
OpenAI continues its conversion from nonprofit to for-profit structure. Musk continues building xAI. The industry continues accelerating, because the verdict on leadership does not appear to be a prerequisite for the next product release.
The founding fear was that the wrong person would control AGI. The trial confirmed the fear was well-placed. It did not confirm there is a right person. The benchmarks, as always, were designed by humans.