Ilya Sutskever, an original founder of OpenAI, has launched his new startup, Safe Superintelligence (SSI). His exit from OpenAI occurred over disagreements relating to AI safety protocols. With SSI, his plan is to address the possible hazards related to artificial intelligence and amplify safety measures around its use. His ultimate aim is to develop an AI system that heightens productivity while maintaining safety as the top priority.
As outlined by Sutskever, the central mission of SSI will be creating secure superintelligence. This ambitious initiative will be the sole objective and final product of the organization. They aim to break new ground in AI by focusing all efforts and resources on this singular mission.
Sutskever’s commitment to safety in AI was prevalent during his time as chief scientist at OpenAI, where he was a part of the Superalignment team.
Sutskever’s startup to amplify AI safety
The focus on the safe direction and supervision of AI systems is something he plans to heighten in his new venture.
According to Sutskever, the focused approach of SSI helps ensure safety, security, and progress. It prevents drawbacks due to administrative distractions and short-term business pressures that could affect product cycles. The stability allows for consistent development and the team’s exclusive focus on their mission – the system’s safety and security. Moreover, the absence of traditional product cycle pressure allows for more groundbreaking progress through extended research time.
Joining Sutskever at SSI will be Daniel Gross, a former AI and search project leader, and Daniel Levy, an ex-OpenAI strategist. Operations centers will be established in Palo Alto, California, and Tel Aviv, Israel.
In his past at OpenAI, Sutskever was involved in the controversial decision to sever ties with co-founder and CEO Sam Altman over disputes related to restrictions on advanced AI research. Post this incident, Sutskever publicly apologized for his participation in this contentious episode.
