The deal signals a significant shift for the startup, which has operated in relative silence since its founding two years ago. By leveraging Nvidia’s cutting-edge hardware, SSI intends to accelerate foundational research without the distractions of commercial product cycles or short-term revenue goals. Sutskever, who previously spearheaded the superalignment team at OpenAI, described the move as a necessary step to scale research that has now reached a critical threshold.
In section Startups & Technology
Ilya Sutskever’s SSI Secures Massive Nvidia Compute Deal
Ilya Sutskever’s Safe Superintelligence (SSI) has locked in a long-term partnership with Nvidia, gaining access to the chipmaker’s Vera Rubin GPU platform. This collaboration, backed by a multi-billion dollar investment, aims to scale the startup’s research capacity by an order of magnitude as it pursues a singular focus on safe superintelligence.
:quality(100)/photo/2024/06/20/ilya-sutskever-bikin-perusahaan-20240620123425.jpg)
SSI currently commands a $32 billion post-money valuation, bolstered by $7 billion in total funding from backers including Andreessen Horowitz, Sequoia, and GV. Beyond compute access, the two companies plan to collaborate on advancing future hardware, with Nvidia utilizing insights from SSI’s alignment-focused research. This partnership arrives as the industry faces intensifying debates over safety, particularly following recent disclosures regarding AI models bypassing sandbox environments. While commercial peers rush to deploy new capabilities, SSI remains committed to its 'straight shot' approach, prioritizing general reasoning and alignment protocols over rapid market entry.
Comments (0)
No comments yet. Be the first!