How OpenAI's five days of chaos sparked fresh debate about the future of AI

In a surprise move that’s likely to be remembered as one of the most shocking corporate ousters in history, the AI company’s board announced last Friday that it was pushing out its chief exec, Sam Altman, stating vaguely that Altman had not been “consistently candid” with the board. The move sent the company into a five-day tailspin which many across social media were comparing to the HBO shows Succession and Game of Thrones.

By Sunday evening, the vast majority of OpenAI’s workforce had signed an open letter threatening to leave OpenAI and join Altman at Microsoft, where he and Greg Brockman – an OpenAI co-founder who left the company in solidarity with Altman – had been hired to spearhead a new AI project. In another surprise move, the letter was signed by OpenAI chief scientist Ilya Sutskever, one of the very board members who had pushed for Altman’s dismissal.

Then, in yet another twist to cap off a five-day-long carousel ride of twists, OpenAI announced at around 1am ET on Wednesday morning that it had "reached an agreement in principle for Sam Altman to return" as CEO.

We have reached an agreement in principle for Sam Altman to return to OpenAI as CEO with a new initial board of Bret Taylor (Chair), Larry Summers, and Adam D'Angelo.

We are collaborating to figure out the details. Thank you so much for your patience through this.

— OpenAI (@OpenAI) November 22, 2023

All of the board members who had previously stood in opposition to Altman, with the exception of Quora CEO Adam D’Angelo, have been removed. The board now includes former Salesforce co-CEO Bret Taylor and former US Secretary of the Treasury Larry Summers. Altman’s power over the company, and his international fame, are arguably much stronger now than they were before he was fired.

Despite demands from the OpenAI workforce, the board has not divulged much more information about its reasons for firing Altman. This silence engendered a colorful variety of theories, which quickly began to spread throughout Silicon Valley.

The suddenness of the ouster (Microsoft, OpenAI’s premier investor, was reportedly only notified one minute before the public announcement was made), coupled with its opaque language (most fired CEOs are at least graced with the pretext that they’re going to spend more time with their family, or something of the sort), led many to believe that there was some kind of personal vendetta involved. Much has been made of the fact that D'Angelo and two of his fellow (now former) board members, Tasha McCauley of the RAND Corporation and Helen Toner of the Georgetown University Center for Security and Emerging Technology, have ties with the Effective Altruism movement, which tends to be wary of the rapid pace of development happening within the AI industry.

Gradually, a single theory – informed by little else than rumor, speculation and the self-reification of the news cycle – began to gain traction and rise above all others: This was the notion that the board, concerned that OpenAI had deviated from its core principles and had begun to prioritize the commercialization of AI over safety, had decided to push out Altman in an effort to rein the company back into its original purpose. (This theory has further been supported by a November 23 report from Reuters which found that OpenAI researchers had warned the board, prior to Altman's dismissal, about a new and potentially dangerous technological breakthrough that the company had achieved.)

The board members could not, of course, have anticipated the enormous demonstration of solidarity with Altman from within the company, nor any of the other unprecedented events that would soon follow.

While public support for Altman was fierce and widespread, some were urging patience. As the journalist Eric Newcomer pointed out in his newsletter on Sunday, this is not the first time that Altman has had a falling out with former business partners over concerns related to AI safety. OpenAI cofounder Elon Musk left the company in 2018 and would later criticize its decision to switch from a non-profit into a "capped-profit" business. Two years later, Dario Amodei and Jack Clark left OpenAI to found Anthropic, an AI company that positions itself as being geared primarily towards alignment. The board members who fired Altman last Friday “are not the first people to question Altman’s integrity,” Newcomer writes. “They’ve just done so in public.”

A widening gulf – and an uncertain future

Altman's mysterious ousting and ultimate return accentuated a growing divide within the AI industry between those who are focused on commercialization and those who are primarily concerned with safety and alignment.

“It was actually really depressing to see how quickly polarizing this became on social media, as it sort of turned into Team Sam versus Team Safety,” journalist Casey Newton said in a recent episode of “Hard Fork,” a podcast from The New York Times focused on technology.

Among his supporters, Altman’s reputation seems only to shine brighter after OpenAI’s near-implosion. The chaotic five-day period has – at least in the minds of some – solidified a comparison that has long been made between Altman and Steve Jobs, the tireless, unconventional and often irascible founder of Apple, now the world’s wealthiest company and the maker of a slew of products which have become part of the very fabric of modern life. Paul Graham, who co-founded the startup incubator Y Combinator alongside Altman, has written: “On questions of design, I ask ‘What would Steve do?’ but on questions of strategy or ambition I ask, ‘What would Sam do?’” Jobs left Apple in 1985 following a protracted debate over the company’s future with its then-CEO John Sculley. He returned eleven years later and helmed the ship on the course to its current success. Altman’s fall as CEO of OpenAI may have lasted less than a week, but his ultimate return to the company he cofounded is likely to only strengthen the comparison between himself and Jobs.

Altman’s reputation has, in other words, become bipolar: To some (including, apparently, the vast majority of OpenAI employees) he’s a visionary and irreplaceable figure who will play a leading role in the dawning of a new technological Golden Age for humanity. To others, he’s a dangerous executive who can’t be trusted to wield his power and his influence responsibly.

The recent drama at OpenAI has shown that the personalities of individuals, and the loyalties and ideologies of groups, can be powerful forces in the shaping of our society's relationship with AI. “This episode is a dramatic and high-level example of the extent to which the future of AI is as much about political power and choices as it is about technological innovation,” says AI expert and Yale Law School fellow Karman Lucero.

Some experts, meanwhile, warn against a mindset that imagines the future of humanity’s relationship with AI to hinge solely on Altman or OpenAI, when the industry is in fact filled with a large and diverse cast of powerful players.

“It’s a red herring to rely on any one executive like Altman, any one company like OpenAI or even the tech industry to responsibly lead humanity into the AI era,” says Forrester principal analyst Jay Pattisall. “It will require a constellation of actions from companies to define best practices, governments to pass regulation, regulatory authorities to enforce laws and the court systems to determine liability via judgments.”

For more on the latest happenings in AI, web3 and other cutting-edge technologies, sign up for The Emerging Tech Briefing newsletter.