Ilya Sutskever

After helping trigger and then reverse Sam Altman’s 2023 ouster, Ilya Sutskever shifted from OpenAI’s governance crisis to building Safe Superintelligence and later testifying about that conflict.

Who they are

Ilya Sutskever appears in coverage as an OpenAI co-founder and former chief scientist whose work and board-level role put him at the center of the company’s 2023 leadership rupture. After leaving OpenAI in 2024, he co-founded Safe Superintelligence with Daniel Gross and Daniel Levy, a startup explicitly focused on safe superintelligence; by July 2025, Sutskever had become its CEO.

The recent arc

Coverage peaked in late 2023 during OpenAI’s attempt to remove Sam Altman: Sutskever was initially identified with the board’s position, then signed the employee letter demanding Altman’s reinstatement. Stories also tied him to OpenAI research, including reporting on a breakthrough he spearheaded, before Altman returned and Sutskever was left outside the reconstituted board.

The tension

The coverage repeatedly circles a conflict between AI-safety-oriented governance and OpenAI’s leadership and commercialization trajectory. Sutskever co-led OpenAI’s Superalignment effort with Jan Leike, warned against dismissing open-source AI as a “side show,” and later emerged in accounts and testimony concerning his concerns about Altman. His creation of Safe Superintelligence turns that unresolved disagreement into a competing institutional path, while Meta’s reported attempt to acquire SSI and recruitment of Gross show how aggressively larger platforms are pursuing the same frontier-talent pool.

Why it matters

Sutskever’s arc matters because it links the governance failure at a leading AI lab with the formation of a new safety-branded frontier-AI company. If SSI can retain technical leadership and financing after Gross’s departure, it could become a meaningful alternative center for work on superintelligence; if not, the episode may instead underline how difficult it is for independent labs to sustain a safety-first mission amid competition from OpenAI, Meta, and other well-capitalized actors.

Related: OpenAI · Sam Altman · AGI · Ilya Sutskever, Daniel Gross, and Daniel Levy announce Safe Superintel · Ilya Sutskever and over 700 out of ~770 OpenAI staffers sign a letter