Departures tracker
Who left OpenAI in 2026
13 recorded departures, newest first, each with the role, the date, a source and a confidence label. The AILately.com Desk adds to this list as the Signal publishes. Last entry 2026-09-16.
Diogo Almeida launches TypeSafe AI with Jev decision model
A ChatGPT co-inventor bets that machine-only decision models beat chatbots at scale for automated tasks.
Fidji Simo joins Nscale board as independent director
Simo carries OpenAI product credibility onto Nscale's board as the infrastructure firm heads toward a US IPO, pairing raw compute ambition with a deployment leader fluent in shipping consumer-facing AI.
Brad Lightcap to Unannounced Startup as Founder
OpenAI's longest-serving operator steps away to found something of his own, leaving a structural void at the top.
Fidji Simo to ChronicleBio as Co-Founder
Simo trades OpenAI's applications helm for a biotech venture built from her own medical journey.
Fidji Simo to ChronicleBio as Founder
Removed the executive most directly responsible for turning OpenAI's research into consumer revenue, timed awkwardly against IPO preparation.
Kate Rouch to unannounced as unannounced
Compounded April's operating leadership loss at OpenAI ahead of Simo's July exit, deepening the pattern CNBC later flagged to investors.
Kevin Weil to unannounced as unannounced
The first of three senior OpenAI operating departures inside a six-month span, preceding the wider exodus narrative.
Shengjia Zhao to Meta as Chief Scientist, Meta Superintelligence Labs
Gave Meta a direct technical line to the model architecture behind OpenAI's flagship product.
Mira Murati to Thinking Machines Lab as Founder and Chief Executive Officer
Created the founder-scientist lab that later drew John Schulman away from Anthropic and became a magnet for OpenAI alumni.
John Schulman to Anthropic as Researcher
Opened a two-hop migration that eventually carried RLHF expertise from OpenAI through Anthropic to Thinking Machines Lab inside ten months.
Daniel Levy to Safe Superintelligence Inc. as Co-Founder
Completed Safe Superintelligence's three-person founding bench alongside Sutskever and Gross.
Ilya Sutskever to Safe Superintelligence Inc. as Co-Founder
Removed OpenAI's most senior technical co-founder and, within a year, produced a $30 billion independent rival.
Jan Leike to Anthropic as Alignment Researcher
Carried core alignment-research expertise out of OpenAI and into Anthropic at the same moment Sutskever was leaving to found SSI.




