I dedicated my life to OpenAI and did not want it to be destroyed, former chief scientist testifies: Altman sowed discord among executives to create chaos

📄Full Article· Automatically extracted by trafilaturaGemini 翻譯1736 words

The lawsuit filed by Musk against OpenAI has entered its final stage. Former Chief Scientist Sutskever testified that Altman undermined executive relationships and pitted them against each other, while also revealing the inside story behind the dissolution of the Superalignment team. (Previous coverage: "I thought he was going to hit me": OpenAI President recounts the 2017 haunted manor breakup night: Musk was too dictatorial) (Background: Musk once demanded 51.2% of shares and wanted to use SpaceX as salary payment — OpenAI founding emails fully exposed) The lawsuit filed by Musk against OpenAI and Microsoft has entered its final stage, with Microsoft CEO Satya Nadella, former OpenAI Chief Scientist Ilya Sutskever, and current Chairman Bret Taylor testifying in court. Sutskever's testimony drew the most attention. He publicly disclosed for the first time that he holds approximately 0.82% of shares in OpenAI's for-profit arm, which is currently valued at about $7 billion based on the company's $850 billion valuation. OpenAI President Greg Brockman, who testified earlier, also admitted that he holds OpenAI shares worth approximately $30 billion. Brockman described himself and Sutskever as having been "inseparable" until November 2023, when Sutskever helped lead the effort that resulted in Sam Altman's brief removal as CEO. (Extended reading: Sam Altman returning as CEO? Microsoft steps in to manage the OpenAI board, both sides enter "closed-door negotiations") According to statements made by lawyers on Monday, Sutskever spent about a year collecting evidence of Altman's dishonest behavior, confirming that Altman's actions "undermined executive relationships and pitted them against each other," and assisted in drafting the memo submitted to the board. Sutskever's testimony reinforced Musk's claim that Altman is unfit to lead an AI lab capable of creating AGI. He mentioned that the Superalignment team he helped lead, which focused on the safety of future models, was "the most important work in the long run" for OpenAI. However, the team was dissolved shortly after Sutskever's departure in May 2024. Sutskever stated: "I felt like I poured my life into it (OpenAI), I just cared about it and didn't want it to be destroyed." However, Sutskever also defended OpenAI, noting that Musk never negotiated any special commitments when funding the OpenAI non-profit. He stated that OpenAI needed "a massive amount of capital" to build computers as large as the human brain, and the transition to a for-profit company was a consensus direction. Musk's legal team had attempted to label Sutskever as a hostile witness due to his financial interests in OpenAI, but the judge rejected this. However, Gonzalez Rogers agreed to give lawyers from both sides greater flexibility in questioning Sutskever, as he is in a "unique position" in the case. Sutskever testified for about an hour, stating that he previously supported firing Altman because the "environment where executives did not have the correct information" was not conducive to "achieving any grand goals," but he criticized his fellow board members for acting in haste, lacking experience, and accepting poor legal advice. In the lawsuit, Musk alleges that Microsoft helped transform OpenAI into a money-making machine that exceeded his original intentions. Microsoft CEO Nadella testified that Microsoft initially provided cloud computing support to OpenAI at a discounted rate, but "once the bills started rising," it became unaffordable. It was easier for Microsoft to accept investing in the for-profit arm in exchange for potential financial returns. As time passed and bills continued to climb, Microsoft wanted more from the partnership. "Microsoft will lose $4 billion next year," Nadella exclaimed in an email to subordinates regarding the OpenAI partnership in 2022. He demanded a new agreement to ensure Microsoft could also gain AI "expertise" from the startup, consistently spelling the company name as "Open AI." Nadella wrote: "If we are going to spend this kind of money and not control our destiny, it makes no sense." Musk's lawyers accused Nadella of pushing OpenAI to generate revenue, showing text messages from early 2023 where Nadella told Altman to launch the ChatGPT paid subscription "as fast as possible," and asked about registration numbers just weeks later. This partnership has been highly effective for Microsoft. As of March 2025, Microsoft has generated $9.5 billion in sales from OpenAI, including an agreement where the research lab shares 20% of its revenue. If OpenAI generates profits, Microsoft's gains will be slightly higher than those of OpenAI's non-profit owners. Nadella argued that this arrangement is fair, considering the risks Microsoft took early on, and noted that if OpenAI were a typical company, Microsoft's claim on profits would have

Data Status✓ Full text extractedRead Original (動區 BlockTempo)

🔍Historical Similar Events· Keyword + Asset Matching6 items

2026-04-30

Sam Altman: OpenAI aims to be a "permanently low-margin" large-scale infrastructure, high profits are simply unrealistic

Similarity 140%關鍵字 altman/openai同分類 zh

2026-04-23

OpenAI CEO Sam Altman slams Anthropic: Fear-mongering Claude Mythos is just to monopolize AI

Similarity 140%關鍵字 altman/openai同分類 zh

2026-05-11

OpenAI’s Next Model Could Be Named “Goblin,” Sam Altman Hints

Similarity 130%關鍵字 altman/openai

2026-05-12