Sam Altman was set to return as OpenAI CEO after five days of upheaval over his firing. The agreement also brought a major change to the company’s board, as pressure from employees and discussions with Microsoft shaped a path back for the executive.
A return after five days of turmoil
OpenAI said it had “reached an agreement in principle for Sam Altman to return to OpenAI as CEO.” The company and Altman were still working out the details, so the announcement described an agreement in principle rather than a finished account of every term.
Altman had accepted a job at Microsoft, a major OpenAI investor. Microsoft also offered to hire OpenAI employees, many of whom threatened to resign en masse if Altman was not brought back. Altman said Microsoft CEO Satya Nadella supported his return and described the companies’ partnership as something he wanted to build on.
OpenAI’s interim CEO, Emmett Shear, said he was pleased with the outcome after what he described as “~72 very intense hours of work.” He said he believed the result balanced safety with the interests of the stakeholders involved.
The board was set for a major reset
The agreement meant three of the four board members who remained after Altman and OpenAI CTO Greg Brockman departed on Friday would leave. Quora CEO Adam D’Angelo was the one remaining member staying on the board. He was reportedly involved in the discussions that led to Altman’s return.
The three departing members were OpenAI Chief Scientist Ilya Sutskever, entrepreneur Tasha McCauley, and Helen Toner of the Georgetown Center for Security and Emerging Technology. OpenAI said its “new initial board” would include D’Angelo, economist Larry Summers, and former Salesforce co-CEO Bret Taylor as chair. More members were expected to join, but Altman would not immediately regain his former board position.
The Wall Street Journal reported that the previous board and Altman’s camp had also agreed to an independent investigation. It would examine Altman’s conduct, the decision to remove him, and the aftermath. The report said the new board could add as many as six additional members.
Why the conflict had become so difficult
OpenAI’s announcement of Altman’s firing gave little detail. It said he “was not consistently candid in his communications with the board, hindering its ability to exercise its responsibilities.” Later reporting described broader tensions, including disagreements about AI safety, the pace of development and the company’s commercialization.
OpenAI has an unusual structure: a nonprofit controls the for-profit subsidiary OpenAI Global. The nonprofit board’s role was described as ensuring that the company develops AI for humanity’s benefit, even if that meant wiping out its investors. That structure placed the board’s mission and the company’s commercial activity in the same story.
According to people familiar with the board’s thinking, the dispute was not attributed to one isolated event. The Wall Street Journal described a gradual loss of trust, along with concerns about Altman’s outside AI-related ventures and how OpenAI technology or intellectual property might be used. Bloomberg also reported that Altman had been seeking money for an AI-focused chip company that would compete against Nvidia.
A fragile resolution under pressure
Altman’s firing quickly became a wider test of OpenAI’s stability. The company’s former executive team pressed the board to reinstate him, warning that its decision could trigger the company’s collapse. Employees’ threat to leave and Microsoft’s offer to hire them added further pressure to resolve the dispute.
The agreement addressed the immediate leadership crisis and set out a new board structure, but it left details to be worked through and an investigation to be conducted. The episode also exposed competing concerns within OpenAI: who should lead the company, how quickly it should develop and commercialize AI, and how its governance should uphold its stated mission.