Safe Superintelligence: Ilya, NVIDIA and a US$32bn AI Bet
Safe Superintelligence CEO Ilya Sutskever is betting a US$32bn AI research lab on a single product, backed by NVIDIA Vera Rubin compute and Google Cloud
TL;DR
- Safe Superintelligence (SSI) has achieved a US$32bn valuation and a major partnership with NVIDIA without releasing any products or research papers.
- The company was founded by Ilya Sutskever, a co-founder of OpenAI, along with Daniel Gross and Daniel Levy, with a sole mission to create a safe superintelligence.
- SSI has raised approximately US$8bn across three funding rounds, with investors including NFDG, Andreessen Horowitz, Sequoia, DST Global, SV Angel, Greenoaks, Lightspeed, and Alphabet.
- A July 2026 partnership with NVIDIA includes a reported US$5bn investment and grants SSI priority access to NVIDIA's Vera Rubin platform, significantly increasing its compute power.
- SSI primarily utilizes Google Cloud's Tensor Processing Units (TPUs) for its research, a less common choice among AI frontier labs.
- The lab maintains a small team of around 50 people, contrasting with larger rival organizations.