OAKLAND, Calif. — Sam Altman shifted in his seat and gave halting responses as he took the witness stand for the first time in the bombshell trial over OpenAI’s future – with Elon Musk’s lawyer grilling him over whether the jury should believe what he says.
“Are you completely trustworthy?” Musk’s lawyer Steven Molo immediately asked on Tuesday as he stood up to begin cross-examining Altman in the second week of the federal trial.
“I believe so,” Altman replied, having taken the witness stand for the first time in a dark suit and tie.
“You don’t know if you’re trustworthy?” Molo shot back, to which Altman jumped in and said, “I amend my answer to ‘yes.’”
Molo continued hammering Altman over his alleged prevarication – a central talking point in Musk’s case – as he ran through a laundry list of witnesses during the past week who have called out Altman over allegedly inconsistent and contradictory statements and behavior.
Altman – who under friendly questioning from his own lawyers earlier had dished allegations that Musk had tried to seize control of OpenAI in its early days – at times stuttered and appeared to be on his heels.
“Do you always tell the truth?” Molo asked sharply.
“I’m a truthful person,” Altman said, somewhat sheepishly.
“That wasn’t my question,” Molo said.
Musk’s lawyer then recounted allegations of Altman not being truthful – including testimony from former OpenAI board members Helen Toner and Tasha McCauley. In taped testimony this week, OpenAI’s former head of technology Mira Murati had accused Altman of “saying one thing to one person and completely the opposite to another person.”
Altman repeatedly said Tuesday he didn’t see their testimony, citing his “very busy day job.”
When Molo asked, “Do you care that people came here under oath and called you a liar?” the exec said he didn’t agree with that characterization.
Musk has slammed Altman alleging he can’t be trusted after he manipulated him into funding OpenAI the non-profit before launching a for-profit entity.
Earlier in the day, Altman hit back at Elon Musk’s claims that he “stole” OpenAI when it was a charity and steered the company away from its non-profit mission – claiming that Musk himself was busy vying to control OpenAI in the first years after it was founded.
Taking the witness stand for the first time in the landmark trial over OpenAI’s future, Altman on Tuesday recounted one “hair raising” moment during OpenAI’s early days – when several OpenAI co-founders asked Musk what would happen to the company if Musk died and he had control of the company.
“Control of OpenAI should pass to my children,” was Musk’s response, Altman told the packed courtroom.
Altman said that anecdote and Musk’s seemingly lax views on AI safety made him “extremely uncomfortable.”
OpenAI lawyer William Savitt asked Altman directly not long after he took the stand how he feels about Musk’s accusation that he “stole a charity.”
“It feels difficult to even wrap my head around that framing,” said Altman, who appeared in Oakland federal court looking relaxed and wearing a dark suit and grey tie. “It does not fit with my conception of stealing a charity.”
Altman also took some politely worded digs at Musk – for instance, Altman described a “moral boost” at the company when Musk became less involved with OpenAI. Altman said Musk’s demands to know what workers were contributing to the company on a short term basis was out of step with the research they were pursuing.
The trial, in its third week, has featured a clash of tech titans that could alter the future of OpenAI and its leadership.
Musk is targeting Altman for removal from OpenAI in a federal lawsuit that claims the company violated its charitable mission. Musk is seeking $180 billion in damages from OpenAI and Microsoft, pledging to donate any proceeds from a court victory to OpenAI’s charitable arm. He is also asking the court to restore OpenAI’s nonprofit status and to remove Altman and Brockman from their roles.
The showdown has gripped Silicon Valley and beyond, with witness testimony at times displaying the personalities and inner workings of the upper echelons of the tech world.
The San Francisco-based maker of hit AI tool ChatGPT has raised hundreds of billions of dollars from large tech companies and investors and could hold a potential trillion-dollar IPO.
Altman used his testimony to highlight OpenAI’s current safety efforts, defend himself from the character attacks that former OpenAI head of technology Mira Murati lodged against him, as well as Helen Toner and Tasha McCauley.
For instance, Murati said in taped testimony played in court that Altman created an environment where OpenAI executives were pitted against each other, creating “chaos” in a way that “undermined” her ability to do her job.
“My concern was about Sam saying one thing to one person and completely the opposite to another person,” Murati said.
Musk’s lawsuit alleges Altman and OpenAI manipulated him into funding the nonprofit with $38 million, before it abandoned its charitable mission to benefit humanity and instead become a for-profit corporation that enriched Altman and others.
OpenAI has argued in court that Musk knew about and at times supported the for-profit plan. The company said Musk walked away from OpenAI after the other co-founders denied his proposals to have majority control.
Former OpenAI chief scientist Ilya Sutskever testified on Monday that he spent about a year gathering evidence for the company’s board demonstrating Altman’s “consistent pattern of lying.”
Musk in court said that he was a “fool” to trust Sam Altman with the future of OpenAI – “I was a fool who provided them free funding to create a startup,” Musk said. “I gave them $38 million of essentially free funding to create what would become an $800 billion company.”
