Ilya Sutskever, co-founder and former chief scientist of OpenAI, has secured $1 billion in funding for his new company, Safe Superintelligence (SSI). Sutskever aims to take a different approach to superintelligence compared to OpenAI.
According to executives who spoke with Reuters, SSI has completed a $1 billion financing round, valuing the company at $5 billion. The funds will be used for computing power and hiring, with SSI CEO Daniel Gross emphasizing their interest in employees who are "not interested in the scene, in the hype."
Investors include prominent venture capital firms such as Andreessen Horowitz, Sequoia Capital, DST Global, and SV Angel, as well as NFDG, an investment partnership led by Nat Friedman and Daniel Gross. Gross, who previously led AI initiatives at Apple, said investors are willing to fund "a couple of years doing R&D" before a product comes to market.
Sutskever hints at new technical approach
SSI chief researcher Sutskever announced a different approach to scaling than OpenAI, without providing specifics. "Everyone just says scaling hypothesis. Everyone neglects to ask, what are we scaling?" he said
"Some people can work really long hours and they'll just go down the same path faster. It's not so much our style. But if you do something different, then it becomes possible for you to do something special."
The "scaling hypothesis" suggests that by simply making AI models larger and training them on more data, they will continue to improve in performance and capability. This approach has led to significant advances in AI, such as GPT-3 and its successors. It's not clear whether, or to what extent, this hypothesis will hold true for future AI systems.
Sutskever left OpenAI in May after being involved in the temporary dismissal of OpenAI CEO Sam Altman as a board member in November 2023. He had reportedly expressed concerns about the rapid commercialization Altman was pushing and the associated security risks.
Unlike OpenAI, which is gradually developing more advanced AI systems with products like ChatGPT, SSI aims to directly develop a "secure super AI" without detours. The startup, founded about three months ago, is based in Palo Alto and Tel Aviv.
