OpenAI co-founder Ilya Sutskever announced Thursday that he will step into the CEO role at Safe Superintelligence, an artificial intelligence startup he founded in 2024, following the recent departure of former CEO and co-founder Daniel Gross. Gross officially left the company on June 29, amid reports that Meta CEO Mark Zuckerberg has been actively courting him for a key position at Meta’s new AI unit. Safe Superintelligence co-founder Daniel Levy has been appointed as the company’s president, according to Sutskever.
In announcing these leadership changes, Sutskever addressed ongoing speculation regarding potential acquisitions. While Meta was reportedly interested in acquiring Safe Superintelligence entirely—at what was said to be a valuation near $32 billion—Sutskever rejected these overtures, clarifying that the company remains independently focused on its core mission.
“We have the compute, we have the team, and we know what to do,” Sutskever wrote in a statement posted online. “Together we will keep building safe superintelligence.”
The departure of Gross, particularly given his reported negotiations with Meta alongside his investment partner Nat Friedman (former CEO of GitHub), naturally raises questions about Safe Superintelligence’s near-term prospects. Specifically, it invites scrutiny as to why a co-founder would choose to step away from such an ambitious project—which positions itself uniquely as the “world’s first straight-shot safe superintelligence lab,” singularly dedicated to the creation of safe and beneficial superhuman AI.
By contrast, Meta appears likely to harness Gross’s leadership in AI product development to power a variety of its consumer offerings, a role closer to Gross’s previous experiences, including leading AI development teams at Apple following the acquisition of his earlier startup. Zuckerberg’s recently announced Meta Superintelligence Labs division emphasizes integration of advanced AI technologies into consumer-facing products, suggesting a strategically distinct direction from Safe Superintelligence’s dedicated focus.
Zuckerberg’s push to enhance Meta’s AI capabilities has already seen the addition of notable researchers previously associated with OpenAI and Google DeepMind, suggesting the ambitious depth and scope of Meta’s investment in artificial intelligence.
For Sutskever, meanwhile, stepping into the CEO role at Safe Superintelligence marks a new career chapter after his pivotal role as OpenAI’s chief scientist. This higher profile, executive-level position may introduce challenges distinct from his former technical leadership responsibilities, including securing continued investor confidence, raising future capital, and recruiting high-level scientific talent. Nonetheless, Sutskever stated clearly that he intends to remain closely involved with the company’s technological direction.
