Andrew Tulloch, one of the prominent researchers caught up in Silicon Valley’s increasingly aggressive AI talent battle, has left Meta after spending roughly 11 months back at the company. His departure comes less than a year after Meta reportedly recruited him following an extraordinary compensation offer that was initially reported to be worth as much as $1.5 billion over six years.
The Wall Street Journal reported that Tulloch informed colleagues of his departure, although his next destination and the reason for leaving have not been disclosed. The development is particularly notable because Tulloch had previously turned down Meta’s reported offer before eventually agreeing to return to the company in October 2025.
Andrew Tulloch Became a Major Target in Meta’s AI Hiring Push
Tulloch has built an impressive career across some of the biggest names in artificial intelligence. He previously worked at Facebook and later joined OpenAI, before becoming one of the co-founders of Thinking Machines Lab, the AI startup established by former OpenAI chief technology officer Mira Murati.
His reputation made him a particularly valuable target as Meta attempted to strengthen its position in the rapidly developing AI industry. The Wall Street Journal reported in 2025 that Mark Zuckerberg personally became involved in efforts to recruit talent from Thinking Machines Lab after an attempt to acquire the startup failed.
The reported compensation package for Tulloch became one of the most eye-catching examples of the AI talent war. The Wall Street Journal said the package could have reached $1.5 billion over at least six years when potential bonuses and equity gains were considered.
However, Meta strongly disputed the characterization. A company spokesperson described the reported figure as “inaccurate and ridiculous,” according to Axios [source]. Other reports also indicated that the eventual package Tulloch accepted was lower than the earlier offer that he had initially rejected.
Tulloch ultimately joined Meta’s newly established TBD Lab, which was part of Zuckerberg’s broader effort to assemble a team capable of competing at the cutting edge of artificial intelligence. The division was being led by Alexandr Wang, the founder of Scale AI who joined Meta during the company’s aggressive AI expansion.
His Meta Return Lasted Less Than a Year
Tulloch’s return to Meta was significant because of his history with the company and his standing within the AI research community. Earlier reporting described him as a highly regarded machine-learning expert, while former Facebook executive Mike Vernal told the Wall Street Journal that Tulloch was “known as an extreme genius.”
His career also highlights how dramatically compensation for elite AI researchers has changed. In 2016, when OpenAI was still in its early stages, Greg Brockman reportedly attempted to recruit Tulloch from Facebook. At the time, Tulloch was earning around $800,000 annually, while OpenAI’s proposed compensation was substantially lower, leading him to remain at Facebook.
Nearly a decade later, the situation was completely different. Meta was competing for Tulloch while attempting to build one of the industry’s most ambitious AI research organizations.
Now, his second stint at Meta has ended after approximately 11 months. Semafor, as reported by Yahoo Finance [source], said Tulloch had delayed his departure until Meta successfully launched its new open-source AI models and Muse AI assistant.
Neither Meta nor Tulloch has publicly explained why he decided to leave. His personal website currently lists him as working at Meta’s TBD Lab on superintelligence, although the latest reports indicate that his departure is now underway.
The move underscores the extraordinary volatility surrounding AI research talent, where even billion-dollar headlines and access to enormous computing resources have not guaranteed that leading researchers will remain at a company for the long term.
