Sam Altman’s OpenAI Ouster: New Testimony Reveals Chaos
▼ Summary
– Ilya Sutskever testified that Sam Altman exhibited a pattern of manipulation, including pitting executives against each other and providing conflicting information about company plans.
– Sutskever compiled a 52-page memo documenting Altman’s behavior and sent it to board members, stating he waited at least a year before proposing Altman’s removal.
– Former board member Helen Toner publicly stated Altman systematically concealed information from the board, including his OpenAI startup fund ownership and ChatGPT’s release.
– During Altman’s brief ouster, Anthropic proposed merging with OpenAI, though discussions were short-lived due to practical challenges.
– Following the leadership turmoil, Sutskever and other key executives left OpenAI to start rival AI companies, while an independent board review later reinstated Altman.
The 2023 removal and swift reinstatement of OpenAI CEO Sam Altman sent shockwaves through the technology sector, raising serious questions about corporate governance and leadership stability at one of the world’s most influential AI companies. New testimony from co-founder Ilya Sutskever reveals the internal chaos and allegations of manipulative behavior that precipitated this dramatic boardroom coup, providing unprecedented insight into the tensions that nearly toppled the organization.
During a nearly ten-hour legal deposition, Sutskever detailed his growing concerns about Altman’s leadership style, which he claims included pitting executives against one another and providing conflicting information about company direction. The testimony forms part of Elon Musk’s ongoing lawsuit against OpenAI and offers the most comprehensive account yet of the events leading to Altman’s brief ouster.
Sutskever described compiling a detailed 52-page memorandum documenting what he perceived as Altman’s pattern of undermining executives and creating internal conflict. He sent this document directly to board members while deliberately keeping it from Altman, explaining that he feared the CEO would “find a way to make them disappear” if he became aware of the discussions. The co-founder acknowledged he had been considering proposing Altman’s removal for at least a year before the actual events unfolded.
The deposition reveals specific instances where Sutskever believed Altman manipulated situations to his advantage. He cited an example involving former OpenAI research executive Dario Amodei, who allegedly demanded control over all research operations and the termination of president Greg Brockman. According to Sutskever, Altman refused to either accept or reject these conditions, seemingly playing both sides to determine which outcome would benefit him most.
Further allegations emerged through claims attributed to former OpenAI CTO Mira Murati, who reportedly provided Sutskever with documentation suggesting Altman had similarly pitted her against then-research executive Daniela Amodei. The testimony also referenced concerns about Altman’s previous leadership role at Y Combinator, where he was allegedly accused of “creating chaos, starting lots of new projects, pitting people against each other.”
These revelations align with previous public statements from former board member Helen Toner, who asserted that Altman systematically withheld critical information from the board. She claimed the board discovered ChatGPT’s public release through Twitter and was unaware of Altman’s ownership stake in the OpenAI startup fund, creating potential conflicts of interest. Toner emphasized that the board reached a point where they “couldn’t believe things that Sam was telling us,” making effective oversight impossible.
The brief period following Altman’s removal brought additional turmoil, including merger discussions with rival Anthropic. Sutskever recalled that both Dario and Daniela Amodei participated in these talks, which ultimately collapsed due to practical challenges. The failed coup resulted in significant board restructuring and eventually led to the departures of several key figures, including Sutskever himself, who left to establish competing AI company Safe Superintelligence.
The legal proceedings continue to uncover new dimensions of this corporate drama, with Sutskever’s testimony representing just one piece of the puzzle. While many questions remain unanswered due to redactions and incomplete records, the deposition provides crucial context for understanding one of the most significant leadership crises in recent technology history. As the legal battle progresses, further disclosures may shed additional light on the internal dynamics that shaped these extraordinary events at the forefront of artificial intelligence development.
(Source: The Verge)
