- Ilya Sutskever, an OpenAI cofounder, has founded a new company.
- Sutskever was involved in a chaotic coup against OpenAI CEO Sam Altman in November.
- On Wednesday, Sutskever said he's starting a new company, Safe Superintelligence Inc.
OpenAI cofounder and former chief scientist Ilya Sutskever has started a new company.
The former OpenAI exec announced on Wednesday that his new venture was a research lab committed to developing "safe superintelligence."
Sutskever left OpenAI in May after months of uncertainty surrounding his role within the AI lab. His departure was swiftly followed by Jan Leike, who led OpenAI's superalignment group, a team that focuses on aligning its AI systems with human interests, with Sutskever.
Following his departure, OpenAI CEO Sam Altman said on X: "Ilya and OpenAI are going to part ways. This is very sad to me; Ilya is easily one of the greatest minds of our generation, a guiding light of our field, and a dear friend. His brilliance and vision are well known; his warmth and compassion are less well known but no less important."
OpenAI president and cofounder Greg Brockman also thanked Sutskever for "being my co-founder, my friend, and the officiant at my civil ceremony" in a separate post.
Speculation about Sutskever's role in ousting CEO Altman has been bubbling for months. The former chief scientist appeared to play a key role in the attempted coup which he later came to regret.
In November, Altman was fired in a shock board decision, kickstarting weeks of chaos for the company.
The move initially shocked the tech community and OpenAI investors, including Microsoft. Reports quickly started circulating that the board was wavering on its controversial decision and actively trying to get Altman back.
After staffers staged a revolt in the form of an open letter demanding the board's resignation and the reinstatement of their former chief, negotiations between OpenAI and Altman appeared to resolve, and the company announced he was set to return as CEO.
Altman's return also came with a new board that included lead investor Microsoft.
Here's a timeline of OpenAI leadership chaos that led us to this point.
Altman was ousted from OpenAI in a shock board decision
Altman was immediately ousted from OpenAI in a dramatic board decision on November 17. The company's nonprofit board announced it no longer had "confidence in his ability to continue leading."
A timeline provided by fellow OpenAI cofounder Greg Brockman showed that Altman received a text from board member Sutskever asking to talk at midday Friday.
When Altman joined the meeting, the whole board, excluding Brockman, was on the call, and he was told by Sutskever he was being fired, per Brockman's post.
OpenAI staffers resigned
Brockman, the president and cofounder of OpenAI, quit his role at the company shortly after the news broke. He said he was told he was being removed from the board and that Altman had been fired at about the same time OpenAI published a blog post about the incident.
Brockman said he was originally told he was "vital to the company and would retain his role," but decided to part ways with the company shortly after.
The tech community was shocked by the decision, and at least three other OpenAI researchers followed the former president in giving up their roles.
OpenAI board was optimistic about getting Altman back
Investors, including Microsoft CEO Satya Nadella, were reportedly blindsided by the decision to remove Altman.
Several investors began pushing the board members to reverse their sudden leadership decision, Bloomberg reported.
In an internal memo cited by The Information, OpenAI's chief strategy officer, Jason Kwon, wrote to staff, saying the company remained optimistic that it could bring back Altman and other senior employees who had left.
Microsoft chief Nadella reportedly led the talks on Altman's return over that weekend.
New CEO announced
The Information later reported that Sutskever told OpenAI staffers that Altman would not return, prompting several more employees to quit.
Discussions to bring Altman back had reportedly fallen apart, per The Verge. The board then named former Twitch CEO and cofounder Emmett Shear as interim CEO. Shear took over the role from OpenAI's CTO, Mira Murati, who had publicly supported Altman.
Altman and Brockman joined Microsoft
In a strange twist, Microsoft CEO Satya Nadella scooped up Altman and Brockman to lead "a new advanced AI research team" at Microsoft.
In a post on X, Brockman said the new Microsoft team would comprise him, Altman, Aleksander Madry, Szymon Sidor, and Jakub Pachocki. He said the team planned to "build something new" and "incredible."
Nadella announced on Monday the company remained "committed to our partnership with OpenAI" and had "confidence in our product roadmap."
Key board member said he regretted ousting Altman
OpenAI cofounder and chief scientist Sutskever said he deeply regretted his involvement in firing Altman.
Many had theorized that Sutskever had a major role in the decision, with reports highlighting the differing priorities between Altman and the nonprofit board.
"I deeply regret my participation in the board's actions. I never intended to harm OpenAI," Sutskever said in an X post following a weekend of chaos. "I love everything we've built together and I will do everything I can to reunite the company."
The chief scientist also signed an open letter calling for the board, which he sits on, to resign.
Employee revolt
Most employees returning to work on Monday were unhappy with the chaos the OpenAI board had kickstarted.
Former CTO Mira Murati was installed as interim CEO, following Altman's departure. CEO Emmett Shear, who cofounded Twitch, then took over from her. This meant the company had three CEOs in as many days.
The majority of OpenAI staffers just wanted Altman back.
By the end of the day, nearly the entire OpenAI staff had threatened to quit and join Altman at Microsoft. They signed a letter calling the board to resign, accusing members of undermining the company.
Altman's return was rumored to still be on the cards
Despite publically joining Microsoft that morning, by the evening, Altman was reportedly still considering returning to his old post at the AI lab.
The Verge reported that Altman and Brockman were considering going back to OpenAI if the board members who ousted Altman stepped down.
Multiple sources told the outlet that Altman, along with Brockman and OpenAI's investors, were trying to find a plan for board members to exit the company.
Altman and Brockman set to return to OpenAI
On November 21, OpenAI confirmed that Altman was set to return as CEO.
The ChatGPT-maker said in a post on X they "reached an agreement in principle" for Altman's return.
Shortly after OpenAI broke the news, Altman reflected on his decision to join Microsoft, saying he thought "it was clear that was the best path for me and the team."
He continued: "With the new board and w satya's support, i'm looking forward to returning to openai, and building on our strong partnership with msft."
Nadella also commented on the move, adding Microsoft was "encouraged by the changes to the OpenAI board."
OpenAI also announced a "new initial board" consisting of Bret Taylor, Larry Summers, and Adam D'Angelo. Taylor, a former Salesforce executive, will serve as board chair.
Microsoft snags a board seat
The next day, Microsoft announced that the company was getting a non-voting board seat.
There had been speculation that Satya Nadella would seek more control over OpenAI following the board's failed coup.
The CEO was reportedly frustrated by the board's lack of communication during Altman's firing and rehiring. Later, he told journalist Kara Swisher he did not plan to be surprised by the company again.
In a message to employees posted on the company's website, re-instated CEO Altman said: "We clearly made the right choice to partner with Microsoft and I'm excited that our new board will include them as a non-voting observer."
A board member speaks out
Former OpenAI board member Helen Toner announced her resignation from the company following Altman's return.
In a post on X, Toner sought to quell rumors the board was attempting to slow some of OpenAI's work.
She said: "To be clear: our decision was about the board's ability to effectively supervise the company, which was our role and responsibility. Though there has been speculation, we were not motivated by a desire to slow down OpenAI's work."
Several reports had pointed to a mysterious new OpenAI model known as Q* as a potential trigger for the chaos. The model is said to have sparked concern at the startup.
Top execs quit months later
In May, two top executives said they had resigned from OpenAI.
Sutskever, one of the company's cofounders, announced he was leaving the company six months after Altman's failed ouster.
The former chief scientist played a key role in the failed coup of OpenAI's CEO. He later appeared to regret his involvement and publicly apologized on social media.
Sutskever said he was confident OpenAI would build "safe" technology. He added that he was excited about his next steps and planned to share more details "in due time."
Jan Leike, the co-lead of OpenAI's superalignment group, also announced his resignation.
Sutskever's new startup
Just over a month after leaving OpenAI, Sutskever announced his new venture: Safe Superintelligence Inc, or SSI.
The announcement was made in an X post, which said the company had "one goal and one product: a safe superintelligence."
It planned to achieve this by advancing its "capabilities as fast as possible while making sure safety always remains ahead."
"This way, we can scale in peace," the post added.
Besides Sutskever, the company also lists former Apple AI lead Daniel Gross and ex-OpenAI technical staff member Daniel Levy among its cofounders.
