OpenAI co-founder and former chief scientist Ilya Sutskever’s new AI startup, Safe Superintelligence (SSI), is utilizing Google Cloud’s TPU chips to energy its AI analysis, a part of a brand new partnership the businesses introduced on Wednesday in a press launch.
Google Cloud says SSI is utilizing TPUs to “speed up its analysis and improvement efforts towards constructing a secure, superintelligent AI.”
Cloud suppliers are chasing a handful of unicorn AI startups that spend lots of of hundreds of thousands of {dollars} on computing energy yearly to coach AI basis fashions. SSI’s cope with Google Cloud suggests the previous will spend a big chunk of its computing price range with Google Cloud; a supply acquainted tells TechCrunch that Google Cloud is SSI’s major computing supplier.
Google Cloud has a historical past of putting computing offers with its former AI researchers, a lot of whom are actually operating billion-dollar AI startups. (Sutskever as soon as labored at Google.) In October, Google Cloud mentioned it could be the first computing supplier for World Labs, based by ex-Google Cloud AI chief scientist Fei-Fei Li.
It’s unclear if SSI has struck partnerships with different cloud or computing suppliers. A Google Cloud spokesperson declined to remark. A Safe Superintelligence spokesperson didn’t instantly reply to a request for remark.
SSI got here out of stealth in June 2024, months after Sutskever departed from his position as OpenAI’s chief scientist. The firm has $1 billion in backing from Andreessen Horowitz, Sequoia Capital, DST Global, SV Angel, and others.
Since SSI’s launch, we’ve heard comparatively little in regards to the startup’s actions. On its web site, SSI says that growing secure, superintelligent AI programs is “our mission, our identify, and our complete product roadmap, as a result of it’s our sole focus.” Sutskever beforehand mentioned that he had recognized “a brand new mountain to climb” and is investigating new methods to enhance the efficiency of frontier AI fashions.
Before co-founding OpenAI, Sutskever spent a number of years at Google Brain researching neural networks. After main OpenAI’s AI security work for years, Sutskever performed a key position within the ousting of OpenAI CEO Sam Altman in November 2023. Sutskever later joined an worker motion to reinstate Altman as CEO.
After the ordeal, Sutskever reportedly wasn’t seen at OpenAI’s workplaces for months, and in the end left the startup to start out SSI.
