Hi, ChatGPT, who is the CEO of OpenAI? The answer, like much of the current inner workings of the world’s top AI startup, has been changing from hour to hour.
You’re not hallucinating: The company has been in a state of utter chaos since the bombshell announcement that its board had fired CEO Sam Altman last Friday afternoon. The twist reportedly caught pretty much everyone but the board members behind the move by surprise, and concerned parties—Microsoft CEO Satya Nadella, OpenAI’s investors and employees—spent the rest of the weekend trying to wrangle some semblance of order out of the situation.
The big question that hasn’t been answered publicly is why any of this is happening. While it’s been reported across multiple outlets that the dispute centered on OpenAI drifting from its original mission as a nonprofit-turned-hybrid org, the board members haven’t said anything in detail beyond claiming that Altman “was not consistently candid in his communications” with them. Indeed, even OpenAI Chief Scientist Ilya Sutskever, reportedly the instigator of the whole conflict, tweeted Monday morning that he “deeply regrets” his role in it, confusing many.
The original personnel announcement, released just last Friday, might feel like weeks ago at this point, given the speed of news since. So here’s a timeline of everything that’s happened between Friday and the publication of this story on Wednesday.
Last Friday, noon PT: As OpenAI’s now-former president Greg Brockman tells it in a tweet, Sutskever texted Altman on Thursday night asking to meet the next day. When Altman joined a Google Meet on Friday at noon, the entire board, sans Brockman, was on the call and told Altman he was fired. That group includes Sutskever, Quora CEO Adam D’Angelo, the Center for Security and Emerging Technology’s Helen Toner, and Tasha McCauley, CEO of city mapping startup GeoSim.
Brockman said he got a text from Sutskever 20 minutes later asking to meet. He jumped on a Google Meet and was told he was being removed from the board but “would retain his position” at the company, he said. Brockman resigned later that evening.
Last Friday, around 12:30pm PT: A blog post from OpenAI made the conflict public, reportedly blindsiding OpenAI’s investors, including Microsoft—which has agreed to invest more than $13 billion in the company—Axios and Forbes first reported. Former CTO Mira Murati was appointed interim CEO. The blog post cryptically claimed that Altman’s lack of candor with the board led to the decision, setting off rampant speculation about what Altman possibly could have done to merit such treatment as a CEO seemingly at the top of his game.
Later Friday afternoon: The Information reported that at an all-hands meeting with OpenAI employees, Sutskever seemed to imply that firing could have been an effort to return OpenAI to its original mission as a nonprofit to develop AI “that benefits all of humanity.”
OpenAI was founded as a nonprofit organization in 2015, but it switched over to a hybrid “capped profit” structure after the release of GPT-2 in early 2019. Since then, it’s gradually become more like a regular for-profit tech company through decisions like not revealing much about the code behind GPT-4. Some in the AI community speculated that the company’s recent DevDay event might have been a wake-up call to proponents of its nonprofit mission.
Keep up with the innovative tech transforming business
Tech Brew keeps business leaders up-to-date on the latest innovations, automation advances, policy shifts, and more, so they can make informed decisions about tech.
It’s not the first time this has come up. One of OpenAI’s chief competitors, Anthropic, was founded by Dario Amodei, who said he left OpenAI because of a belief it wasn’t sufficiently concerned with AI safety.
Friday evening: Brockman announced his resignation, and The Information (whose team had excellent real-time reporting of the weekend’s events) reported that three senior researchers had resigned.
Saturday morning: The Information reported that an $86 billion employee share sale could be endangered by the internal turmoil.
Saturday evening: As investors reportedly pressured the board to reinstate Altman, per the New York Times, OpenAI employees showed their support for the former CEO by quote-tweeting one of his posts with a heart emoji en masse.
Sunday: Altman reportedly returned to OpenAI headquarters to negotiate his return on the condition that the board that fired him step down, The Verge reported. He set a 5pm PT deadline for a new deal.
Sunday night: In a shock decision, OpenAI named Twitch co-founder and former CEO Emmett Shear as its next interim CEO. In addition to his experience at the video streaming platform, internet users and media dug up some of Shear’s more colorful tweets and role as a character in a “seminal Harry Potter fanfiction” supposedly popular with members of the effective altruism movement, 404 Media reported. Shear said in a tweet that he would hire an independent investigator to prepare a report that would ”dig into the entire process leading up to this point” within 30 days.
Monday morning: Microsoft hired Altman and Brockman to lead a new AI research lab within the company. Hundreds of OpenAI staffers, including Sutskever, penned a letter threatening to quit and join Microsoft unless Altman and Brockman are reinstated and the board resigns, Wired reported. As of Monday evening, 95% of OpenAI’s roughly 770 employees had reportedly signed the letter.
Tuesday morning: Microsoft Chief Technology Officer Kevin Scott tweeted that the company would offer a role with matched compensation to any OpenAI employees leaving for Altman’s new research lab at Microsoft.
Tuesday, 10pm PT: After five days of drama, OpenAI announced that Altman will return as CEO under a new board. The initial board will consist of Salesforce co-CEO Bret Taylor, who will serve as chair; former Treasury secretary and Harvard President Larry Summers; and one holdover, Quora CEO Adam D'Angelo.
“We are collaborating to figure out the details,” the announcement said. “Thank you so much for your patience through this.”
While things may seem to be back to roughly where they were when this all started last Friday—give or take a big-name economist—the episode could shake confidence in the many companies that are building their AI on OpenAI models. As more information emerges about what happened, some are questioning AI safety governance, especially when billions are on the line.
Update 11/22/23: This piece has been updated to include news that broke after the story was first published Tuesday afternoon.