Safe Superintelligence Inc. (SSI), a new company founded by Ilya Sutskever, has emerged just a month after his departure from OpenAI. Sutskever, who served as OpenAI’s chief scientist, launched SSI alongside former YC partner Daniel Gross and ex-OpenAI engineer Daniel Levy. This move comes after Sutskever and Jan Leike, another key figure in OpenAI’s AI safety efforts, left the company due to disagreements with leadership regarding AI safety approaches. Leike now heads a team at Anthropic.
Sutskever has long been focused on the complex aspects of AI safety. In a blog post published in 2023, he and Leike predicted that superintelligent AI systems surpassing human intelligence could be developed within the next decade. Recognizing the potential risks associated with such advancements, they emphasized the need for research into controlling and restricting AI.
SSI’s main objective is to build safe superintelligence, viewing it as the most crucial technical problem of our time. The company aims to establish the world’s first dedicated SSI lab, with a singular focus on developing a safe superintelligence. The company’s name, mission, and entire product roadmap revolve around this goal.
In a tweet, SSI stated that they approach safety and capabilities concurrently, perceiving them as technical challenges that can be addressed through groundbreaking engineering and scientific breakthroughs. They plan to push the boundaries of AI capabilities while ensuring that safety remains a top priority. This approach allows them to scale their operations without being hindered by management overhead or short-term commercial pressures.
Currently, SSI has offices in Palo Alto and Tel Aviv and is actively recruiting technical talent. This expansion indicates their commitment to assembling a skilled team capable of tackling the challenges associated with developing safe superintelligence.
The emergence of SSI underscores the growing recognition of the importance of AI safety and the need for dedicated efforts to ensure the responsible development of superintelligent AI systems. With Sutskever’s expertise and the team’s single-minded focus, SSI is poised to make significant contributions to the field, advancing the understanding of AI safety and paving the way for a safer and more controlled future of artificial intelligence.
