At a Glance
- Thinking Machines co-founders Barret Zoph and Luke Metz rejoin OpenAI days after Zoph’s firing
- Zoph was terminated for “unethical conduct” after allegedly leaking confidential data to rivals
- Half of Thinking Machines’ co-founders have now left the $12 billion startup within months
- Why it matters: The talent shuffle highlights cut-throat competition for AI expertise and raises questions about loyalty and IP safety in the sector
The AI talent war turned personal this week when two Thinking Machines Lab co-founders jumped back to OpenAI immediately after one was ousted for ethics violations. Barret Zoph, the startup’s former CTO, was fired on Wednesday following accusations he shared sensitive information with competitors. Roughly 60 minutes later, OpenAI announced Zoph and fellow co-founder Luke Metz were rejoining its ranks.
$2 Billion Startup Loses Top Talent
Thinking Machines, launched in 2025 by ex-OpenAI CTO Mira Murati, closed a $2 billion seed round over the summer, drawing investment from Andreessen Horowitz, Nvidia, and AMD. The deal valued the company at $12 billion and produced Tinker, an API for fine-tuning large language models.
The company’s leadership roster is now shrinking fast. Andrew Tulloch departed for Meta last fall, and with Zoph and Metz gone, half of the original co-founders have exited. Sources told News Of Fort Worth that additional staffers are in late-stage talks to follow them to OpenAI.
Timeline of a Rapid Exit
| Date | Event |
|---|---|
| Monday | Zoph informs Murati he is considering leaving |
| Wednesday | Zoph is terminated for “unethical conduct” |
| Wednesday, ~1 hour later | OpenAI announces Zoph, Metz, and researcher Sam Schoenholz are rejoining |
Wired obtained an internal memo from OpenAI applications CEO Fidji Simo stating that Zoph’s recruitment had been “in the works for several weeks.” Simo also wrote that OpenAI does not share Murati’s concerns about Zoph’s behavior.
The Backstory Behind the Breakup
Murati, Zoph, and Metz all previously worked together at OpenAI. Murati served as interim CEO during Sam Altman’s brief removal in 2023. The Wall Street Journal reported she supplied Slack screenshots that board members used to justify concerns about Altman’s leadership. After Altman returned, Murati resumed her CTO post before announcing in 2024 she would step away “to create the time and space to do my own exploration.”
Tech reporter Kylie Robinson first revealed Zoph’s termination, citing unnamed sources inside Thinking Machines. Murati confirmed the departure in a post on X, and Simo followed with her own post welcoming the trio back to OpenAI: “This has been in the works for several weeks, and we’re thrilled to have them join the team.”
A Revolving Door Across AI
The incident underscores how frequently researchers swap employers in the AI sector. Meta reportedly spent millions last year to recruit staff from OpenAI, only to see some researchers return within weeks. The practice has become so common that contracts, valuations, and even product roadmaps can hinge on which team a handful of specialists joins.
Neither OpenAI nor Thinking Machines responded to Cameron R. Hayes‘s requests for comment before publication.
Key Takeaways
- Thinking Machines has lost three of six co-founders in less than a year
- The exits come just months after a headline-grabbing $2 billion raise
- OpenAI gains experienced talent amid ongoing competition for AI supremacy
- Allegations of leaked confidential data add a layer of controversy to the hiring
