Ilya Sutskever Launches Safe Superintelligence Inc. to Focus on AI Safety

In a bold and timely move, Ilya Sutskever, one of the co-founders of OpenAI and its former chief scientist, has launched a new venture called Safe Superintelligence Inc. (SSI). This comes merely a month after his departure from OpenAI, marking a significant shift in the landscape of artificial intelligence development and safety research. Sutskever’s new company, co-founded with former Y Combinator partner Daniel Gross and ex-OpenAI engineer Daniel Levy, aims to tackle one of the most pressing challenges of our time: ensuring the safety of superintelligent AI systems.

A New Mission: Safe Superintelligence

Sutskever, a pivotal figure in AI research, was instrumental in OpenAI’s efforts to enhance AI safety, particularly as the development of superintelligent systems accelerated. During his tenure, he collaborated closely with Jan Leike, who co-led OpenAI’s Superalignment team. Their joint efforts were directed towards mitigating the risks associated with superintelligent AI, a field that Sutskever has long been passionate about. However, both Sutskever and Leike left OpenAI in May following disagreements with the leadership on the approach to AI safety. Leike has since joined Anthropic, another leading AI research organization.

Predicting the Future of AI

Sutskever’s commitment to AI safety is well-documented. In a 2023 blog post co-authored with Leike, he predicted that AI with intelligence surpassing human capabilities could emerge within the decade. This foresight underscores the urgency of developing mechanisms to control and restrict such AI to prevent potential harm. The formation of SSI reflects Sutskever’s unwavering dedication to this cause.

The Vision of Safe Superintelligence Inc.

SSI’s mission is clear and singular: to create a safe superintelligence. The company’s name, Safe Superintelligence Inc., embodies its primary objective and product focus. According to a tweet announcing the company’s formation, SSI is dedicated to advancing AI capabilities while ensuring that safety measures remain ahead. This approach allows for scalable development without compromising security.

“Our mission, name, and entire product roadmap revolve around SSI,” the announcement stated. “We approach safety and capabilities in tandem, solving technical problems through revolutionary engineering and scientific breakthroughs. Our goal is to advance capabilities rapidly while maintaining safety, enabling us to scale peacefully.”

Business Model and Approach

Unlike OpenAI, which started as a non-profit before restructuring to meet the financial demands of AI development, SSI is designed from the outset as a for-profit entity. This strategic decision reflects the evolving landscape of AI research, where substantial investment is required to drive progress. Daniel Gross emphasized this point in a conversation with Bloomberg, indicating that fundraising will not be a hurdle for SSI given the intense interest in AI and the impressive credentials of its team.

Conclusion

The launch of Safe Superintelligence Inc. by Ilya Sutskever marks a significant milestone in the AI industry. By focusing exclusively on developing safe superintelligent systems, SSI aims to address one of the most critical technical challenges of our era. With a clear mission, a dedicated team, and a robust business model, SSI is poised to make substantial contributions to the field of AI safety. As Sutskever and his team embark on this ambitious journey, the world watches with keen interest, anticipating the advancements and breakthroughs that will shape the future of artificial intelligence.