Ilya Sutskever Defends His Role in Sam Altman’s OpenAI Ouster: ‘I Didn’t Want It to Lose’

It’s Elon Musk a case against OpenAI and Microsoft entered its final phase on Monday, with evidence from Microsoft CEO Satya Nadellaformer OpenAI chief scientist Ilya Sutskever, and OpenAI chairman Bret Taylor.

Sutskever expressed interest, revealing ownership of OpenAI’s for-profit arm of $850 billion which is currently worth about $7 billion. This makes it one of the most popular OpenAI domains. Earlier in the case, OpenAI President Greg Brockman made the first admission that he owns about $30 billion in shares of OpenAI.

Brockman was one of the founders of the investigation, and Sutskever joined soon after, refusing compensation of $ 6 million per year from Google. Brockman said he and Sutskever were “joined at the hip,” until Sutskever helped lead Sam Altman’s team. temporarily removed as CEO of OpenAI in 2023. Sutskever helped gather evidence showing Altman’s fraudulent history, and also helped write a memo to the board. Despite attempts to mend the relationship, Sutskever has been estranged from Brockman and Altman since then, OpenAI’s attorney said Monday.

Sutskever, who appeared in court in a dress shirt and slacks, the first male witness to testify without a suit jacket, appeared distraught at the loss of his association with OpenAI. (He left and made competitive AI lab in 2024.) “I felt like I had a lot of ownership of OpenAI,” he said at one point on Monday. “I felt like I put my life in it, and I just took care of it, and I didn’t want it to be destroyed.”

Sutskever’s testimony bolstered Musk’s argument that Altman is not the right person to lead an AI lab capable of creating artificial intelligence. In addition, Sutskever mentioned how the superalignment team he helped leadwhich focused on the security of future models, was playing a key role in OpenAI “for a long time.” The group was disbanded in May 2024, shortly after Sutskever left the company.

But Sutskever added to OpenAI’s defense that Musk did not discuss specific commitments to fund the nonprofit OpenAI. Musk’s claims that his promises were in place and that Altman and Brockman broke them in pursuit of the for-profit arm are the basis of the lawsuit. Sutskever said OpenAI needs “a lot of dollars” to build a supercomputer like the human brain, and while seeking donations has had “reasonable success,” becoming a for-profit is the way forward.

“I can tell the difference between an ant and a cat,” Sutskever said in response to a question from US District Judge Yvonne Gonzalez Rogers about how computers support OpenAI. “If there is no money, there is no supercomputer.”

Finally, Sutskever, a famous AI scientist who paints in his spare timehe testified for about an hour, not making eye contact with anyone during the time he was on the witness stand.

Musk’s legal team was unsuccessful in getting Sutskever as a malicious witness because of his investment in OpenAI. But Gonzalez Rogers agreed to give Musk’s and OpenAI’s lawyers more time to question Sutskever because of what he described as his “special position” in the case.

The Blip

Much of Monday’s testimony focused on the well-covered events surrounding Altman’s dismissal and reinstated as CEO in November 2023. Nadella described Sutskever and the other members of the board who fired Altman as “city fans” and reiterated that he “didn’t feel good” about the lack of justice that led to their decision. Nadella also admitted in his testimony that he and his colleagues discussed it 14 potential board members who will join OpenAI if Altman returns, including at least two who were voted on by the Microsoft team and one who joined later. Nadella described Microsoft’s proposal as a proposal.

Sutskever said he supports firing Altman because “an environment where directors don’t have accurate information” is not “ideal for any purpose.” But he criticized his colleagues in the committee for rushing the project, lack of skills and accepting “legal advice that was not very good.”

Bet for Microsoft

In his lawsuit, Musk accused Microsoft of helping turn OpenAI into a more money-making machine than Musk wanted. Nadella testified that Microsoft initially supported OpenAI with low-cost cloud computing but could not do so “once the bill started to mount.” The profit arm that Microsoft could invest in, to get a return on investment, was very sweet.

But as the years went by and the bills piled up, Microsoft demanded more from the deal. Microsoft “will lose 4 bil next year!!!” Nadella said in an email in 2022 to its sponsors of the OpenAI collaboration. He called for a new agreement to ensure that Microsoft also gets “knowing” AI back to basics, which he labeled “Open AI.”