Ilya Sutskever, co-founder of OpenAI, has launched Safe Superintelligence Inc. (SSI), a company dedicated to developing safe superintelligence. After leaving OpenAI in May 2024, Sutskever announced SSI on X (formerly Twitter), emphasizing its focus on creating a secure superintelligence without immediate commercial plans.

SSI aims to tackle cybersecurity through innovative engineering and scientific advancements. The company, based in Palo Alto and Tel Aviv, involves key collaborators like Daniel Gross and Daniel Levy. Sutskever, known for his deep learning expertise, expressed enthusiasm for SSI's mission and invited interested individuals to join the project.

 

Mission and Vision of Safe Superintelligence Inc.

SSI describes itself as the first lab focused on developing a secure superintelligence. According to its website, the lab's sole objective is to create AI that meets stringent cybersecurity requirements. They aim to advance capabilities while ensuring safety is paramount, addressing security and capability issues as technical problems to be solved through revolutionary engineering and scientific breakthroughs.

In an interview with Bloomberg, Sutskever highlighted that SSI's initial and sole product will be a secure superintelligence, a technology that does not yet exist. He emphasized that SSI will remain free from external pressures and avoid competitive races until achieving its primary goal. The company's founding team includes notable figures like Daniel Gross, former AI lead at Apple, and Daniel Levy, ex-OpenAI employee. Their specific roles within SSI are yet to be disclosed. Also, SSI's announcement includes an invitation for individuals interested in contributing to this ambitious project. Although specific vacancies were not detailed, potential collaborators are encouraged to reach out via the provided email.

 

Legacy and Future Prospects

Ilya Sutskever, renowned for his deep learning work, co-founded OpenAI in December 2015 with prominent figures including Elon Musk and Sam Altman. His involvement in Altman's dismissal in 2023 led to significant internal turmoil at OpenAI. Reflecting on this, Sutskever expressed deep regret and a desire to reunite the company. Upon resigning from OpenAI in May 2024, he conveyed excitement for his new venture and received praise from Altman for his brilliance and compassion.

OpenAI, founded in 2015, has been a milestone in AI research, emphasizing safe and beneficial AI development. The organization has promoted open, collaborative research, aiming for advancements that benefit humanity. Sutskever's creation of Safe Superintelligence Inc. signifies a new chapter in his career and AI's evolution, with a focused yet ambitious mission.