OpenAI will bring back Sam Altman and overhaul its board to bring on new directors including Larry Summers, a stunning reversal in a drama that’s transfixed Silicon Valley and the global AI industry.

Altman is returning as chief executive officer and the initial board will be chaired by Bret Taylor, a former co-CEO of Salesforce.com Inc. and director at Twitter before it was acquired by Elon Musk. Other directors include Summers, the former US Treasury Secretary under President Bill Clinton, and existing member Adam D’Angelo, the co-founder and CEO of Quora Inc. OpenAI is now working “to figure out the details,” the company said in a post on X, formerly Twitter.

OpenAI decided to reinstate its co-founder after nearly all of its employees threatened to quit over his surprise ouster. The post triggered swift congratulations on X from key players in the saga including former president Greg Brockman — who said he too is returning to the fold — and CTO Mira Murati.

Altman, who was fired by OpenAI’s board on Friday following disagreements over how fast to develop and monetize artificial intelligence, had been in negotiations with the company to return. Those talks reached an impasse on Sunday in part over pressure from Altman and others for existing board members to resign, according to people familiar with the matter.

Read More: OpenAI Leaders Tell Staff ‘Get Back to Shipping’ Amid Tumult

Instead, the board named a new leader — former Twitch CEO Emmett Shear — and OpenAI’s biggest backer, Microsoft Corp., said it would hire Altman to head up a new in-house AI team.

Within hours, most of OpenAI’s 770 employees signed a letter to the board saying they might quit and join Microsoft unless all board directors resigned and Altman was reinstated. Among the many who signed the letter was Murati, who had been named interim CEO on Friday, and Ilya Sutskever, an OpenAI co-founder and board member who had previously disagreed with Altman over the company’s direction.

“I deeply regret my participation in the board’s actions,” Sutskever wrote in a post on X, formerly Twitter, on Monday. “I never intended to harm OpenAI. I love everything we’ve built together and I will do everything I can to reunite the company.”

Altman’s return caps a roller-coaster few days that threw the world’s best-known AI startup into chaos. Summers, a Harvard University professor and paid contributor to Bloomberg TV, is on the board of Jack Dorsey’s Block Inc. and also Skillsoft Corp.

The quick reversal could appease investors and reduce the threat of employees fleeing. But it also raises questions about the path ahead for the ChatGPT maker and other AI startups, which have tried to balance developing artificial intelligence responsibly alongside the need to raise vast amounts of capital from investors to support the expensive computing infrastructure required to build these tools.

Founded in 2015, OpenAI was initially established as a nonprofit aimed at advancing AI in a way that would benefit humanity and not be dictated by financial gain. The group later reorganized itself as a capped for-profit entity, raising billions from Microsoft and other investors — with Altman being instrumental to those deals — but it continued to be overseen by a non-profit board. That tension exploded into full view in recent days.

The startup’s investors were blindsided by Altman’s removal. Microsoft, which backed the startup with a more than $10 billion stake, had only a few minutes’ advance notice about Altman’s firing. The software giant began working with investors including Thrive Capital and Tiger Global Management to bring him back, according to people familiar with the matter who asked to remain anonymous discussing private information. When that effort failed, Microsoft agreed to hire Altman and others from OpenAI.

OpenAI also faced new uncertainty over plans to let employees sell their shares at an $86 billion valuation. Thrive, which was expected to lead a tender offer for employee stock, had not yet wired the money as of Saturday and told OpenAI that Altman’s departure would affect its actions, people said.

Read More: OpenAI Investors, Led By Thrive, Angle to Bring Back Altman

More than any other figure, Altman, 38, emerged as the face of a new era of artificial intelligence technology, thanks to the viral success of ChatGPT. Altman was at the center of the industry’s efforts this year to work with regulators and he met regularly with world leaders, including US President Joe Biden and UK Prime Minister Rishi Sunak. On Thursday, he appeared on a panel at the Asia-Pacific Economic Cooperation conference, attended by other executives and world leaders, to discuss the future of AI and its risks.

Behind the scenes, however, Altman clashed with members of his board, especially Sutskever, over how quickly to develop generative AI, how to commercialize products and the steps needed to lessen their potential harms to the public, people with knowledge of the matter have said. OpenAI’s other board members at the time included D’Angelo; Tasha McCauley, CEO of GeoSim Systems; and Helen Toner, director of strategy and foundational research grants at Georgetown’s Center for Security and Emerging Technology.

Alongside rifts over strategy, board members also contended with Altman’s entrepreneurial ambitions. He has been looking to raise tens of billions of dollars from Middle Eastern sovereign wealth funds to create an AI chip startup to compete with processors made by Nvidia Corp., according to a person with knowledge of the investment proposal. Altman was courting SoftBank Group Corp. chairman Masayoshi Son for a multibillion-dollar investment in a new business to make AI-oriented hardware in partnership with former Apple designer Jony Ive.

Altman’s side ventures added complexity to an already strained relationship with the board.

In a statement Friday, OpenAI said Altman’s departure came after an internal review by the board found the chief executive “was not consistently candid in his communications with the board, hindering its ability to exercise its responsibilities.” As a result, it said, “the board no longer has confidence in his ability to continue leading OpenAI.”

The boardroom drama had echoes of other coups in Silicon Valley history. Apple co-founder Steve Jobs was fired as CEO in 1985 only to return more than a decade later. Twitter co-founder Dorsey was pushed out in 2008 and came back as CEO seven years later.

--With assistance from Dina Bass, Ashlee Vance and Ed Ludlow.

(Corrects to show Summers was at Treasury during the Clinton administration)