Larry Summers Is OpenAI’s Surprise Pick to Mend Fences

As Sam Altman looks to steer OpenAI following the turmoil of his ouster and return as chief executive, he will be working with an initial board of directors consisting of two new faces—and one holdover from the previous board that fired him. 

The two joiners are seasoned tech veteran Bret Taylor, who was formerly co-CEO of Salesforce and chairman of Twitter, and Larry Summers, the onetime Treasury secretary and Harvard University president whose appointment surprised some observers. Remaining on the board is Adam D’Angelo, a former Facebook executive and the founder of the question-and-answer website Quora, a potential sign that Altman may not have everything his own way after his dramatic reinstatement late Tuesday.

The appointment of the new directors came after an impasse over who would lead the company that has become synonymous with the boom in artificial-intelligence technology. The confusion over OpenAI’s future and leadership created a dayslong drama that gripped Silicon Valley and the business world. 

Since the board unexpectedly fired Altman on Friday, investors have been pushing the directors to reinstate him, employees have threatened to quit en masse and Microsoft, OpenAI’s largest backer, said it was hiring him to lead a new advanced AI research team.

---

 

 

---

In Summers, OpenAI is tapping a well-known economist and Washington, D.C., whisperer who has said he is a believer in the power of artificial-intelligence capabilities.

“More and more, I think ChatGPT is coming for the cognitive class,” Summers wrote on X in April, predicting it will change the way doctors, financial traders, authors and editors work. He said those in white-collar professions are likely to see more disruption than those who do manual labor.

Summers served as secretary of the Treasury under President Bill Clinton from 1999-2001 and could be called upon for his expertise on navigating Washington’s political landscape as AI comes under greater scrutiny. He has worked in the world of startups before and is on the board of Twitter co-founder Jack Dorsey’s payments company, Block.

U.S. lawmakers witnessing the rise of a powerful new technology with the potential to cause unprecedented harm have been kicking around the idea of forming a federal agency to regulate technology platforms including AI systems.

Summers resigned as president of Harvard in early 2006 after a rocky five years. In 2005, the university’s arts-and-sciences faculty gave him a vote of no confidence after he told a conference on workforce diversity that innate gender differences could help explain why fewer women achieve high-level academic careers in science and math.

---

OpenAI said late Tuesday that Sam Altman will return to the artificial-intelligence company as CEO, with a new initial board. WSJ reporter Tom Dotan analyzes the latest developments. Photo: John G Mabanglo/Shutterstock

---

Taylor, whom OpenAI said would be chairman, joined business-software titan Salesforce in 2016 and served as co-CEO alongside Chairman Marc Benioff until stepping down this year.

Taylor has backed several startups and worked with some of the highest-profile leaders in tech, serving as chairman of Twitter for about a year before Elon Musk bought the company. 

In February, Taylor said he was launching an AI startup with a longtime veteran of Google to try to ride the next great wave of technological innovation.

“Rarely do you encounter a new technology so powerful that it feels inevitable that it will change the course of every industry,” Taylor said in a post on LinkedIn at the time.  

D’Angelo joined OpenAI’s board in 2018. “I continue to think that work toward general AI (with safety in mind) is both important and underappreciated, and I’m happy to contribute to it,” he wrote on X at the time.

He started Quora in 2009 after working for about two years as Facebook’s technology chief, and is now working on his own AI startup, called Poe. The service, which launched publicly in February, describes itself as a platform “designed for seamless conversational experiences, enhanced productivity, and creative content generation.”

Quora in 2017 said it had raised $85 million in Series D funding co-led by Altman and Collaborative Fund, a deal that gave the site a valuation at the time of nearly $1.8 billion.

OpenAi’s announcement on X gave few other details about the board’s direction, saying in its post that the parties were “collaborating to figure out the details.” The board is expected to add as many as six additional members, according to a person familiar with the matter, in addition to continuing its duties.

Emmett Shear, the former CEO of Twitch who had been named as one of two interim chief executives of OpenAI since Friday, said after his appointment he would hire an independent investigator to review the events leading to the upheaval at the company and would push for governance changes if necessary.

After Altman’s return, Shear said on X that he was deeply pleased and “glad to have been a part of the solution.”

The departing board members include Helen Toner, director of strategy and foundational research grants at the Center for Security and Emerging Technology; Tasha McCauley, an adjunct senior management scientist at Rand Corp.; and Ilya Sutskever, the chief scientist of OpenAI.

Toner, who previously worked at Open Philanthropy, a nonprofit with ties to the Effective Altruism movement, clashed with Altman after penning an academic paper saying OpenAI’s safety practices in releasing ChatGPT were inferior to those of its rival Anthropic.
 

Late Tuesday, Toner posted on X, “And now, we all get some sleep.”

Sutskever is a respected AI researcher who made major contributions to the field of “deep learning” while working with Geoffrey Hinton, often regarded as the father of AI. His wooing away from Google to co-found OpenAI in 2015 was a major coup for the then-nascent nonprofit research lab.

More recently, he had been overseeing OpenAI’s “superalignment” team, charged with ensuring that the artificial general intelligence that OpenAI was pressing toward would function in the best interests of humanity.

It was Sutskever who delivered the news Friday that the board had decided to fire Altman. Days later, amid a swell of backlash from OpenAI investors and employees, Sutskever switched sides, joining the list of employees threatening to leave unless Altman was reinstated.

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

Write to Newley Purnell at newley.purnell@wsj.com and Keach Hagey at Keach.Hagey@wsj.com