Sam Altman is back as OpenAI CEO after a rough week. OpenAI’s new board replaced most of the board that tried to oust Altman before Thanksgiving.
Altman announced in a letter circulated internally at OpenAI and published on the OpenAI blog that Mira Murati, who was briefly appointed interim CEO by the previous board, will return to her role as CTO and that the initial new board will include Bret Taylor, the former Salesforce co-CEO; Quora CEO D’Angelo, who served on the previous board; and economist and political veteran Larry Summers.
A non-voting observer from Microsoft will join the board. Microsoft owns 49% of the for-profit OpenAI entity that a nonprofit board controls. This observer won’t vote on board business, but their identity is unknown.
“I’ve never been more excited about the future,” Altman wrote. I’m grateful for everyone’s hard work in an uncertain and unprecedented situation, and I think our resilience and spirit set us apart in the industry.”
Altman outlines OpenAI’s future priorities in the letter, including advancing its research plan and “further investing” in AI safety. Altman promises that the initial board will build a board of “diverse perspectives,” make unspecified “improvements” to OpenAI’s governance structure, and oversee an independent review of recent events.
“It’s important that people get to experience the benefits and promise of AI and shape it,” Altman said. We still think great products are the best. I will work with [OpenAI leadership] to emphasize our commitment to users, customers, partners, and governments worldwide.”
The old board—Alman, OpenAI chief scientist Ilya Sutskever, OpenAI president Greg Brockman, tech entrepreneur Tasha McCauley, D’Angelo, and Helen Toner, director at Georgetown’s Center for Security and Emerging Technologies—suddenly fired Altman without telling anyone, including most of OpenAI’s 770-person workforce. The move infuriated Microsoft and OpenAI’s investors, threatened the company’s rumored stock sale, and caused most employees, including Sutskever, to quit unless Altman was reinstated immediately.
According to reports, Altman and the previous board disagreed on OpenAI’s direction. The board publicly accused Altman of “not being consistently candid” with them. Altman was said to have privately criticized Toner for a paper she co-authored that criticized OpenAI’s safety approach and frustrated Sutskever by rushing AI-powered features at DevDay, OpenAI’s first developer conference.
Altman stated on X (formerly Twitter) that D’Angelo has always been clear about potential conflicts and has taken steps to avoid conflicted decision-making, addressing reports that may have led to his removal. Some see Quora’s Poe chatbot-aggregating service as a competitor to OpenAI.
“We expect that if OpenAI is as successful as we hope, it will touch many parts of the economy and have complex relationships with many other entities in the world, resulting in various potential conflicts of interest,” Altman wrote. We will disclose this and let the board decide how to handle situations like these.