Former OpenAI CTO Testifies CEO Sam Altman Lied About Model Safety in Musk‑Altman Trial
Key Points
- Mira Murati testified that Sam Altman lied about a new GPT model’s safety clearance.
- Altman claimed legal counsel cleared the model without board review; Murati found that false.
- Chief strategy officer Jason Kwon confirmed the model required board oversight.
- Murati said Altman’s management style made her job harder and undermined her authority.
- Earlier accusations from Ilya Sutskever and Helen Toner also described Altman as deceptive.
- OpenAI’s board fired Altman in 2023 for lacking candor; Murati served briefly as interim CEO.
- Murati left OpenAI in 2024 and founded Thinking Machines Lab, a competing AI venture.
In a video deposition aired during the Musk v. Altman lawsuit, Mira Murati, OpenAI’s former chief technology officer, said CEO Sam Altman falsely told her that the company’s legal team had cleared a new GPT model for release without review by the deployment safety board. Murati confirmed a clash with chief strategy officer Jason Kwon over the model’s safety status and said she forced the board review herself. Her testimony adds to earlier accusations from co‑founder Ilya Sutskever and former board member Helen Toner that Altman routinely misled executives and the board, leading to his 2023 dismissal.
During Wednesday’s hearing in the high‑profile Musk v. Altman case, Mira Murati, who served as OpenAI’s chief technology officer, took the stand and directly accused CEO Sam Altman of lying about the safety review process for a new GPT model. Murati recalled Altman telling her that the company’s legal department had determined the model did not need to go before OpenAI’s deployment safety board. When asked whether Altman’s statement was truthful, she answered simply, “No.”
Murati explained that after the conversation she reached out to Jason Kwon, OpenAI’s general counsel turned chief strategy officer, to verify the claim. She found a stark “misalignment” between what Kwon was saying and what Altman had asserted. To protect the organization, Murati insisted the model undergo the board’s safety review, overriding Altman’s directive.
The deposition also shed light on broader tensions within OpenAI’s leadership. Murati described her role as “incredibly hard” in a “very complex” organization, alleging that Altman’s management style hampered her ability to do the job. She said she repeatedly asked Altman to lead with clarity and not undermine her authority. The CTO’s testimony echoed earlier criticisms from co‑founder Ilya Sutskever, who, in a 52‑page memo to the board, accused Altman of a “consistent pattern of lying, undermining his execs, and pitting his execs against one another.” Former board member Helen Toner similarly warned that executives had presented evidence of Altman’s manipulative behavior.
OpenAI’s board, which dismissed Altman in November 2023, cited “a lack of consistent candor” as the primary reason for removing him. In a statement, the board said Altman’s communications “hindered its ability to exercise its responsibilities.” When pressed by The Verge about the board’s loss of trust, Altman replied, “That will be a better question for them.”
Murati’s testimony also touched on the brief period she served as interim CEO after Altman’s ouster. While she acknowledged the board’s decision, she warned that the company faced “catastrophic risk of falling apart” amid the leadership turmoil. She left OpenAI later in 2024 and announced the formation of Thinking Machines Lab, a new venture positioned as a rival to OpenAI.
The courtroom drama underscores the growing scrutiny of AI firms’ internal governance. As the Musk‑Altman trial unfolds, Murati’s statements may influence how courts view corporate accountability in the fast‑moving artificial‑intelligence sector. Whether the deposition will sway the broader legal battle remains to be seen, but it adds a vivid, first‑hand account of discord at the heart of one of the world’s most influential AI companies.