Ex-OpenAI Board Member Accuses Sam Altman of Repeated Dishonesty Before Ouster
Former OpenAI board member Helen Toner claims Sam Altman frequently withheld information and lied to the board, leading to his brief firing last November. She states the board even learned about the launch of ChatGPT on Twitter rather than from company leadership.
Former OpenAI board member Helen Toner breaks her silence to accuse CEO Sam Altman of a years-long pattern of dishonesty. She tells The TED AI Show that Altman withholds information, misrepresents company happenings, and sometimes lies outright to the board. This behavior steadily erodes trust and makes it impossible for the board to fulfill its oversight duties.
Toner dismisses rumors that the board fired Altman due to fears over a sudden technological breakthrough. Instead, she points to specific instances of deception, revealing that the board learns about the famous ChatGPT launch in November 2022 simply by reading Twitter. She explains that Altman consistently offers innocuous-sounding excuses to downplay these serious concerns.
The situation ultimately reaches a breaking point after Toner co-authors a paper favoring Anthropic's safety approach over OpenAI's. Toner states that Altman responds by lying to other board members in an attempt to remove her, which destroys any remaining trust and prompts the brief but dramatic firing. Altman returns to power days later after employees threaten a mass resignation.