Recently Former Chief Scientist of OpenAI, Ilya Sutskever announced his new AI venture, Superintelligence (SSI). The startup, which has been funded with $1 billion from big investors such as Andreessen Horowitz, Sequoia Capital, DST Global, and SV Angel, will build ultra-safe AI systems. Based in Palo Alto, California, and Tel Aviv, Israel, SSI was earlier named Safe Superintelligence.
The investment of $1 billion will be put in a series of ambitious plans to revolutionize the industry. A big backing like that pinpoints the faith investors have in Sutskever's vision for ultra-safe AI technologies. That financial clout would give SSI a free hand in pursuing wide-ranging research and AI development to set new standards in the industry.
Sutskever, famous for his work at Open AI, including the scaling hypothesis behind ChatGPT and other technologies, is looking to solve AI's next frontier. SSI is said to be working on "safe superintelligence," a goal contrasting with the more orthodox approaches at OpenAI. That would mean the new company will be designing AI with safety and ethics in mind, taking the lead in the industry on these two important issues.
The founding of SSI represents the first big move by Sutskever since he left OpenAI, where he was head of the Superalignment group with Jan Leike. He has since already joined rival AI firm Anthropic. Sutskever's departure came amidst much controversy at OpenAI, which included the temporary ousting of its CEO, Sam Altman, for which Sutskever apologized publicly. SSI is Sutskever's next big move to advance AI technology.
Sutskever's leaving OpenAI was wrapped in controversy, particularly about the leadership change within the organization. His new venture reflects a continued commitment to advancing AI. How SSI approaches the line of work in AI safety is very different from the commercial pressures exerted on OpenAI through deep research and ethical development of AI.
Headquartered in Palo Alto and Tel Aviv, SSI underscores its international scope and strategic position within the global landscape of technology. This will foster collaboration and innovation across major technology hotspots.
This move is in line with a growing need for AI ethics and regulations felt by governments worldwide. For instance, the center-left government in Australia has tried to make AI safe through the new regulations, emphasizing human oversight and transparency. The Industry and Science Minister, Ed Husic, introduced 10 voluntary standards for AI to safeguard the deployment of AI. A month-long consultation will be considered, which may make these guidelines binding on high-risk applications. Read More
In light of this, Sutskever will have SSI set a new bar for safety and scaling methodology in AI. The company aspires to do extensive research in developing an AI system that is going to be one of the best performers but with a difference: this time, much emphasis is placed on safety and ethics. This may change how AI technologies are defined and put into practice.