Ilya Sutskever, who helped to create OpenAI, launches AI startup Safe Superintelligence Inc.

Ilya Sutskever, a prominent researcher and co-founder of OpenAI, disappeared from public view after participating in the ouster of OpenAI CEO Sam Altman and then reinstating him in 2023. In mid-May, Sutskever announced his resignation from OpenAI, promising to talk about his next project “in due course.”

Now, Sutskever has unveiled a startup, Safe Superintelligence Inc, which aims to develop a safe and powerful artificial intelligence system as part of a purely research-based effort, with no immediate commercial goals, Bloomberg reports. “This company is special because its first product will be a safe superintelligence, and it will not do anything else until then,” Sutskever said. The researcher plans to create a revolutionary artificial intelligence system isolated from the pressure of commercial competition.

Sutskever did not disclose Safe Superintelligence’s financial backers or the amount of money raised to date.

Safe Superintelligence Inc. has two co-founders: Daniel Gross, an investor and former head of artificial intelligence at Apple, and Daniel Levy, who worked with Sutskever at OpenAI and is known for his expertise in training large artificial intelligence models. The company will have offices in Palo Alto, California, and Tel Aviv, Israel.

Sutskever’s departure from OpenAI and his future plans have sparked intense interest in Silicon Valley. Known for his significant contributions to the development of artificial intelligence as a university researcher and later at Google, Sutskever played a key role in OpenAI’s early success by advocating for the development of large AI models, which contributed to the emergence of technologies such as ChatGPT.

Safe Superintelligence Inc. is somewhat similar to the original concept of OpenAI, a research organization dedicated to creating artificial general intelligence (AGI) that surpasses human capabilities. However, unlike OpenAI, which worked closely with Microsoft and created revenue-generating products, Safe Superintelligence is going to focus exclusively on research and development without the immediate goal of commercializing its AI achievements.

This startup is a significant risk for investors who are betting on the ability of Sutskever and his team to achieve a breakthrough that could put them ahead of competitors with more resources. The term “superintelligence” in the company’s name refers to an artificial intelligence system that would surpass human intelligence, a goal that remains speculative and highly ambitious.