Stay informed with free updates
Simply log in to Artificial Intelligence myFT Digest – delivered straight to your inbox.
OpenAI co-founder Ilya Sutskever is launching a rival AI startup aimed at “building secure superintelligence”, just a month after leaving the AI ​​company following a failed coup attempt against its CEO Sam Altman.
On Wednesday, Sutskever, one of the world’s most respected AI researchers, launched Safe Superintelligence (SSI) Inc, which bills itself as “the world’s first SSI lab with one goal and one product: safe superintelligence.” to the statement published on X.
Sutskever co-founded the breakaway US startup with former OpenAI employee Daniel Levy and AI investor and entrepreneur Daniel Gross, who worked as a partner at Y Combinator, the Silicon Valley startup incubator Altman ran.
Gross has stakes in companies like GitHub and Instacart and AI companies including Perplexity.ai, Character.ai and CoreWeave. Investors in SSI have not been disclosed.
The trio said the development of safe superintelligence – a type of machine intelligence that could replace human cognitive abilities – was the company’s “sole focus”. They added that it will not be burdened by investors’ revenue demands, such as calls for top talent to join the initiative.
OpenAI had a similar mission to SSI when it was founded in 2015 as a non-profit research lab with the goal of creating super-intelligent AI that would benefit humanity. While Altman says this is still a core principle of OpenAI, the company has transformed into a fast-growing business under his leadership.
Sutskever and his co-founders said in a statement that SSI’s “single focus means no distraction from management overhead or product cycles, and our business model means safety, security and progress are insulated from short-term commercial pressures.” The statement added that the company will be headquartered in both Palo Alto and Tel Aviv.
Sutskever is considered one of the world’s leading artificial intelligence researchers. He played a large role in OpenAI’s early leadership in the nascent field of generative artificial intelligence—the development of software that can generate multimedia responses to human queries.
The announcement of its OpenAI offshoot comes after a period of turmoil at the AI ​​front group centered on clashes over leadership and security.
In November, OpenAI’s directors — which at the time included Sutskever — ousted Altman as CEO in a sudden move that shocked investors and employees. Altman returned a few days later under new management, without Sutskever.
After the failed coup, Sutskever remained with the company for several months, but left in May. When he resigned, he said he was “excited about what’s coming next – a project that is very important to me personally and that I will share in due course.”
This is not the first time that OpenAI employees have split from the maker of ChatGPT to create “secure” AI systems. In 2021, Dario Amodei, the former head of AI security at the company, spun off his own start-up, Anthropic, which raised $4 billion from Amazon and hundreds of millions more from venture capitalists at a valuation of more than $18 billion. , according to people familiar with the discussions.
Although Sutskever has said publicly that he has confidence in OpenAI’s current leadership, Jan Leike, another person who recently quit and worked closely with Sutskever, said his disagreements with the company’s leadership “reached a tipping point” as “security culture and processes. they receded into the background of shiny products”. Joined rival OpenAI Anthropic.