Ex-OpenAI exec details year-long probe into Sam Altman

Former OpenAI chief scientist Ilya Sutskever testified in court on Monday that he spent about a year collecting evidence that CEO Sam Altman had shown what he described as a “consistent pattern of lying.”

Speaking during the high-profile legal battle between OpenAI and Elon Musk, Sutskever said he had been considering removing Altman from leadership long before the dramatic boardroom crisis in November 2023, News.Az reports, citing Reuters.

According to his testimony, OpenAI’s board asked him to prepare a detailed document outlining concerns about Altman’s conduct. Sutskever confirmed the report included allegations that Altman undermined executives and created divisions among senior leadership.

He also said he discussed the possibility of removing Altman with former OpenAI CTO Mira Murati after lengthy conversations about the company’s internal tensions.

Sutskever testified that Altman’s behavior was “not conducive to any grand goal,” including the safe development of artificial general intelligence, or AGI.

The testimony revisits one of the most chaotic moments in Silicon Valley’s recent history. In November 2023, Sutskever helped lead the board decision that temporarily removed Altman as CEO. Days later, amid employee backlash and fears the company could collapse, Sutskever publicly expressed regret and supported Altman’s reinstatement.

The ongoing lawsuit could reshape the future of OpenAI, which is pursuing massive fundraising efforts and a potential trillion-dollar IPO. Musk, who co-founded OpenAI before leaving in 2018, accuses the company of abandoning its nonprofit mission in favor of profit and investor enrichment.

Musk is seeking $150 billion in damages from OpenAI and its major backer Microsoft. He is also pushing for Altman and OpenAI president Greg Brockman to be removed from their leadership roles.

During Monday’s testimony, Sutskever also revealed that his OpenAI stake was worth roughly $7 billion. He confirmed that after Altman’s removal, OpenAI board members explored a possible merger with rival AI company Anthropic, though he personally opposed combining the companies.

Sutskever left OpenAI in 2024 and later launched his own startup, Safe Superintelligence.

The courtroom battle has already featured testimony from major AI figures, including Satya Nadella, Murati, Brockman, and Shivon Zilis. Closing arguments are expected later this week.

News.Az 

By Aysel Mammadzada