Summary: Sam Altman testified on May 12, 2026 in Elon Musk’s suit alleging that OpenAI broke its founding agreement. Musk seeks the removal of Altman and Greg Brockman, $134 billion for OpenAI’s nonprofit, and an unwind of the for-profit conversion. The verdict matters because a jury finding could reshape OpenAI’s control, valuation, governance, and the API ecosystem built on it.
The federal courthouse in Oakland hosted Sam Altman as its central witness on May 12, 2026, answering questions in Elon Musk’s suit alleging OpenAI broke its founding agreement. Week three of the jury trial now ends with closing arguments due Thursday, May 14, and the two sides have boiled every dispute down to a single question: did OpenAI’s conversion from nonprofit to for-profit betray its founding mission — or was it, in Musk’s telling, a plot to “steal a charity”?
The remedies Musk seeks are drastic: Altman and Greg Brockman removed from their positions, $134 billion sent back to OpenAI’s nonprofit, and the for-profit conversion unwound. Should the jury grant all of it, the legal footing of OpenAI’s control, valuation, and governance would be rebuilt, and the ecosystem of products built on its APIs would absorb the shock.
The Plaintiff’s Story: A Charity “Stolen”
Steven Molo, the attorney arguing Musk’s case, keeps the narrative spare: quarterly donations from Musk sustained the nonprofit OpenAI, which then turned into a for-profit he would never control — the founding agreement broken. On the stand, Brockman recalled how Musk “stormed around the table” before walking out in 2018, and the record adds a Musk text message, sent just days before trial, threatening Brockman.
OpenAI’s position is that nothing improper occurred and that the suit amounts to jealousy-driven harassment. What lifts the case above a press war is that a jury must find facts: whether the founding agreement existed, whether the conversion betrayed it, and which side has told the truth.
Altman’s Direct Testimony, Then Molo’s Cross
Facing the theft framing, Altman answered without hedging: “We created, through a ton of hard work, this extremely large charity, and I agree you can’t steal it. Mr. Musk did try to kill it, I guess. Twice.” Once Musk’s donations ended, he added, OpenAI ran on an “extremely short runway of cash,” and Musk had done “huge damage for a long time to the culture of the organization.”
Control formed the second thread. Altman showed the jury a 2017 email he had written to Shivon Zilis: “I am worried about control. I don’t think any one person should have control of the world’s first AGI.” Putting the succession question to Musk drew the reply that “maybe control should pass to my children,” which Altman called hair-raising. And of Musk’s overtures to recruit him to Tesla: “I read a vague, like, a lightweight threat in there.”
The cross-examination turned into a trial of character. Molo needed more than ten minutes to enumerate the people who have called Altman a liar: Sutskever, Murati, Toner, McCauley, the Amodei brothers, colleagues from his Loopt days, a New Yorker profile, and the book The Optimist. The Verge’s courtroom read was blunt — “fairly shoddy lawyering from Musk’s side” — with the Stanford fundraising angle blowing up on the questioner, the jury appearing to warm to Altman, and his key statements resting on contemporaneous documents, a prop the testimony from Musk, Zilis, and Brockman mostly lacked.
OpenAI Alumni Weigh In on the “Pattern of Lying”
Musk’s most potent material came from OpenAI’s own former leadership. In a videotaped deposition, Mira Murati accused Altman of “creating chaos” and of “saying one thing to one person and completely the opposite to another person.” Texts from 2023 shown in court captured Altman asking whether his prospects were “directionally good or bad”; Murati’s answer: “Directionally very bad.” Ilya Sutskever appeared in person and confirmed what he had told the board — Altman “exhibits a consistent pattern of lying, undermining his execs and pitting his execs against one another.”
Former board member Helen Toner described a “pattern of behavior related to his honesty and candor,” and Natasha McCauley pointed to “repeated crisis events.” Satya Nadella, Microsoft’s CEO, also testified: no one ever gave him a specific reason for the 2023 firing, he called the board’s handling “sort of amateur city,” and he feared employees would leave en masse.
What Rides on the Verdict
Three things give the outcome weight. First, the nonprofit-to-for-profit structure much of the frontier-lab world has copied or considered — Anthropic and xAI among them — is being publicly judged, and every lab weighing hybrid governance receives a jury’s read on how that conversion fares in litigation. Second, if the $134 billion redistribution request is credited, OpenAI’s capital structure and its control arrangements would have to be rebuilt, leaving developers and product teams that depend on OpenAI APIs facing a legal entity — the one behind their pricing, contracts, and enterprise agreements — that could be swapped out beneath them. Third, timing sharpens everything: The Verge points to a WSJ report that Altman’s investments face political scrutiny ahead of a prospective OpenAI IPO, so the reputational verdict may arrive long before the legal one. Governance and restructuring have been the AI industry’s running theme of 2026; this trial puts that theme under oath.
When the closings conclude, the call belongs to the jury. Either way, the fight over who controls an AGI company now unfolds in a courtroom rather than a boardroom.