Sam Altman returns as CEO with a new initial board
The three-person interim board — Bret Taylor, Larry Summers and Adam D'Angelo, the only holdover from the board that fired him — replaced the four who had voted him out.
- Labs & people
- Major
Sam Altman returned as OpenAI’s chief executive, five days after the board had removed him without giving a detailed public reason. OpenAI announced an “initial” new board of Bret Taylor, formerly co-chief executive of Salesforce, as chair; Larry Summers, the economist and former US Treasury secretary; and Adam D’Angelo, chief executive of Quora, the sole member retained from the board that had voted Altman out. The other three — Ilya Sutskever, Helen Toner and Tasha McCauley — left the board, though Sutskever remained at OpenAI as chief scientist. Greg Brockman, who had resigned as president in solidarity with Altman during the crisis, also returned.
The board gave no detailed public account of why it had acted, and none was subsequently disclosed. The reversal followed pressure from multiple directions: Microsoft, OpenAI’s largest investor, announced within a day of the firing that it would hire Altman and Brockman to lead a new research team regardless of what happened at OpenAI; more than 700 of the company’s roughly 770 employees signed a letter threatening to resign and join them unless the board resigned and reinstated Altman; and Sutskever, who had voted for the removal, publicly said he regretted his part in it.
OpenAI described the new board as “initial,” and it was expanded further later that month, with Microsoft afterward taking a non-voting observer seat. The episode left OpenAI’s unusual governance structure — a non-profit board legally empowered to override commercial considerations, sitting above a capped-profit company reliant on a single dominant investor and a workforce holding valuable equity — as the central fact of the crisis: the board had used the power the structure gave it, and discovered within days that the structure could not sustain the decision. The question of whether that governance model could survive contact with the company’s commercial success ran on through OpenAI’s move toward a more conventional for-profit structure in the following two years.