Altman claims Musk wanted “total control”

TL;DR: It was Sam Altman’s turn on the stand in Silicon Valley’s hottest new legal drama, Musk vs. Altman, yesterday. The OpenAI CEO claimed that Elon Musk was fine with OpenAI becoming a for-profit—as long as Musk had “total control.” Musk’s lawyers sought to paint Altman as a habitual liar who may have misled the Senate about his own equity stake in the company. The defense continues today, with closing arguments expected tomorrow.

What happened: Altman used his time on the stand to go on the offense: Musk wanted full control of OpenAI, had floated folding it into Tesla, and, in one “hair-raising moment,” had suggested passing control of OpenAI to his children when he died, Altman alleged. He claimed that Musk “felt very strongly that if we were going to form a for-profit, he needed to have total control over it initially.”

Musk also allegedly hurt morale during his time at OpenAI with his severe managerial tactics, Altman said; he insisted on ranking researchers by accomplishments and then taking “a chainsaw through a bunch” of them, a strategy that may be familiar to employees at Tesla and the federal government under Musk’s DOGE stint.

“Are you completely trustworthy?” That’s what Musk’s lawyer, Steven Molo, asked Altman flat-out in his cross-examination, and it was a big theme of Molo’s questioning. Molo pointed to previous testimony from other OpenAI insiders who have referred to Altman as untruthful, including former CTO Mira Murati and former board member Tasha McCauley. Molo also tried to bring up Ronan Farrow’s recent New Yorker profile of Altman that centers largely on his untrustworthiness, but the judge quickly shut it down.

Molo also suggested Altman misled the Senate in 2023—testifying he had no equity stake in OpenAI when, it now appears, he did. Altman admitted yesterday that he has an indirect stake through startup incubator Y Combinator’s investment. Altman’s lawyers have since filed a legal brief accusing Molo of rolling out “a parade of unsubstantiated allegations” during the cross-examination.

What’s really at stake: Whether or not Musk can actually win the lawsuit, which could effectively mean the end of OpenAI as we know it, his lawyers are committing to the public record—using testimony from Altman’s former colleagues—assertions that Altman is a liar and someone not to be trusted. Allegations discussed in the trial have already prompted a political probe from the House Oversight Committee, and Republican attorneys general are calling for an SEC review of Altman’s investments. All of this comes just as OpenAI is said to be preparing for a blockbuster IPO.

Bottom line: Both sides have landed substantive blows in this trial, but the question most relevant to OpenAI’s future—whether Altman can be trusted to steer the ship—could continue to plague the company even if OpenAI fends off Musk’s challenge. —PK