SAN FRANCISCO - Sam Altman, who was fired on Friday from his role at ChatGPT-maker OpenAI, will return to his post as chief executive, ending a boardroom drama that has transfixed Silicon Valley and exposed the power struggles over who has control over the future of artificial intelligence.

OpenAI will have an initial board consisting of Bret Taylor, the former chair of Twitter's board before its takeover by Elon Musk; Larry Summers, the Clinton-era U.S. treasury secretary; and Adam D'Angelo, the chief executive of Quora and one of the board members who voted to oust Altman.

"We have reached an agreement in principle for Sam to return to OpenAI as CEO," OpenAI said in a post on the social media platform X, formerly known as Twitter. "We are collaborating to figure out the details. Thank you so much for your patience through this."

Altman's return puts a cap on nearly a week of chaos at the San Francisco-based start-up at the center of the AI boom. His firing on Friday shocked many in the tech industry, including OpenAI's investors and employees, who mounted a campaign to get him back.

On Sunday, Altman went into the OpenAI offices for negotiations, with the aim of securing his return. Employees came in on the weekend to support him, staying for hours and ordering a spread of McDonald's chicken nuggets and fries for sustenance as the evening wore on. But the initial talks broke down and the board announced a new interim CEO. Altman agreed to join Microsoft instead.

That announcement spurred nearly all of the company's roughly 770 employees to sign a letter in protest, threatening to quit if he was not reinstated. Both Altman and Microsoft chief executive Satya Nadella signaled that they would still be open to him going back to OpenAI. On Tuesday, with the threat of the company's chaos mounting, the board reached a new deal with Altman. OpenAI's announcement came around 10 p.m. Pacific time. Jubilant employees gathered in San Francisco's Mission District to celebrate, toasting the return of their leader, according to a person familiar with the situation.

As part of the new deal, Altman will not have a seat on the board, and the board agreed to an independent investigation that will examine all aspects of recent events, including Altman's role, according to a person familiar with the matter, who spoke on the condition of anonymity to discuss sensitive matters. Altman was fired, with the board saying only that he had not been candid in his discussions with board members. On Wednesday, The Washington Post reported Altman had also been fired from a previous job, as head of start-up incubator Y-Combinator, four years ago.

The initial OpenAI board of three will vet and appoint a formal board of up to nine members, another person familiar with the matter said. Ilya Sutskever, Helen Toner and Tasha McCauley, the three other board members who along with D'Angelo moved to oust Altman, will leave the board, the person said. Emmett Shear, who was named interim CEO, will also leave the company, said the person familiar with events around Altman's return.

Greg Brockman, who quit in solidarity with Altman on Friday, said on X he is returning to the company.

A spokesperson for OpenAI declined to comment further.

"I love OpenAI, and everything I've done over the past few days has been in service of keeping this team and its mission together," Altman said on X. "When I decided to join [Microsoft] on sun evening, it was clear that was the best path for me and the team. With the new board and [with] Satya's support, I'm looking forward to returning to OpenAI, and building on our strong partnership with [Microsoft]."

OpenAI shot into the mainstream after releasing ChatGPT, a viral AI chatbot that responds to questions with complex, humanlike responses. The technology has driven a boom in advanced generative AI - illustrating how the technology, which ingests troves of data to create text, images and audio, could transform industries and impact modern life. The software helped propel the company into the center of the AI industry, leapfrogging tech titans such as Google and Meta, which have been working on similar technology.

The drama around Altman's sudden ouster at OpenAI has exposed the deep rift inside the company over who should control its future. OpenAI started in 2015 as a nonprofit research lab, but in recent years under Altman's leadership, it took on billions of dollars in investment from the likes of tech giant Microsoft and venture capitalists, and began developing consumer products. Outside critics and some employees worried the company had abandoned its mission and was behaving more like a Big Tech company when it was originally meant to provide a more transparent, democratic alternative to the tech giants.

The question of who will ultimately control OpenAI and its technology now falls to the three board members. It's unclear whether OpenAI's overall structure and official mission of creating humanlike machine intelligence to benefit all of humanity will be changed as well. Still, a larger board and the presence of Taylor and Summers suggest the future makeup will be more like a traditional tech company, with high-powered, veteran business leaders from different companies, rather than the previous board, which featured Toner and McCauley, who are experts in how AI may impact society, but haven't led giant moneymaking corporations.

Summers has been vocal about artificial intelligence, warning that it is likely to lead to major disruption in the job market much earlier than many forecasters anticipate. He serves on the board of Block, the payments firm started by Twitter co-founder Jack Dorsey, as well as the software company Skillsoft Corp, which put him in contact with numerous Silicon Valley investors. Sheryl Sandberg, the former chief operating officer of Facebook, began her career working with Summers at the World Bank and in the Clinton administration. A spokeswoman for Summers did not immediately return a request for comment on how he joined the OpenAI board.

On Wednesday, the Revolving Door Project, a left-leaning watchdog group that tracks government appointments and their corporate ties, blasted OpenAI's decision to add Summers to its board, criticizing his record of "energetically" embracing the embattled cryptocurrency industry.

"There is no greater indication that OpenAI is unserious about the interests of humanity than their elevation of Larry Summers to its Board of Directors," Jeff Hauser, the group's executive director, said in a statement. Hauser said Summers's "ascent to the heights of AI should accelerate concerns that AI will be bad for all but the richest and most opportunistic amongst us."

Taylor, the other new board member, is no stranger to high-stakes corporate wrangling. Taylor led a successful legal battle to force Musk to go through with the deal to acquire Twitter, a move that was very profitable for Twitter investors, but was criticized by its employees who did not want to work for Musk, and fans of the service who thought the billionaire would destroy it.

Regardless of the final board composition, Altman's return will be greeted with relief by investors, customers and employees, who feared the boardroom drama could lead to the company's implosion. If that had happened, it would have left a vacuum at the center of the AI industry, opening the window for competitors like Google and AI start-up Anthropic to gain momentum.

Microsoft, which is OpenAI's biggest investor and uses its technology in its own AI products, also stands to benefit from a stabilized OpenAI and the return of Altman.

"We are encouraged by the changes to the OpenAI board. We believe this is a first essential step on a path to more stable, well-informed, and effective governance," Nadella said in a statement on X. "We look forward to building on our strong partnership and delivering the value of this next generation of AI to our customers and partners."