Meta’s attempt to poach top AI talent from Mira Murati’s Thinking Machines Lab with compensation packages reportedly ranging from $200 million to over $1 billion ultimately failed, as multiple founding members chose to remain with the startup or return to OpenAI instead. Despite offering what would have been some of the largest individual payouts in tech history, Meta was unable to secure the allegiance of key researchers who cited mission alignment as a primary factor in their decisions. This outcome underscores the growing importance of purpose-driven work in the fiercely competitive AI talent market, where financial incentives alone are no longer sufficient to guarantee retention.
The situation unfolded against a backdrop of intense recruitment activity by Meta, which has been aggressively rebuilding its AI division following internal restructuring. According to verified reports, Meta hired five founding members of Thinking Machines Lab after Murati rejected a reported $1 billion acquisition offer for the company. The most publicized hire was co-founder Andrew Tulloch, who joined Meta in October 2025 with a compensation package reportedly worth $1.5 billion over six years—a figure that, if accurate, would represent the most expensive individual talent acquisition in technology industry history. This move was part of a broader strategy that included Meta’s $14.3 billion investment in Scale AI, the appointment of Alexandr Wang as chief AI officer, and the creation of Meta Superintelligence Labs.
Despite these high-value overtures, not all members of Thinking Machines Lab’s founding team accepted Meta’s offers. In January 2026, three key researchers—Brett Zoph, Luke Metz, and Sam Schoenholz—returned to OpenAI, where they had previously worked before joining Murati’s venture. Fidji Simo, OpenAI’s CEO of Applications, announced the hires on social media, stating that Zoph would report to her while Metz and Schoenholz would report to Zoph. Simo noted that the recruitment had been “in the works for several weeks,” and Bloomberg News reported that Zoph had informed Murati of his intention to leave on a Monday, leading to his termination the same day due to what Thinking Machines Lab described as “unethical conduct.” Neither Zoph, OpenAI, nor Thinking Machines Lab provided comment on the matter when contacted by tech publication Core Memory.
The departures highlight a growing trend in which AI professionals prioritize institutional mission and research freedom over pure financial compensation. Thinking Machines Lab, which was founded by Murati after her departure as OpenAI’s CTO, had raised $2 billion at a $12 billion valuation in a seed round led by Andreessen Horowitz in July 2025. By November 2025, the company was reportedly in talks for a new funding round that could have valued it at $50 billion. Despite this strong financial backing and the allure of Meta’s massive offers, a significant portion of the founding team ultimately chose to remain aligned with Murati’s vision or return to OpenAI, suggesting that cultural and intellectual fit played a decisive role.
Meta’s recruitment efforts have not been without internal costs. The company reportedly cut 600 positions from its Fundamental AI Research (FAIR) team as part of its broader AI reorganization. Yann LeCun, Meta’s longtime chief AI scientist, departed after 12 years with the company. These shifts coincide with the launch of Meta Superintelligence Labs’ first closed-source model, Muse Spark, which debuted on April 8, 2026. The lab was established as part of Meta’s push to develop advanced AI systems capable of reasoning and planning at human levels—a goal that directly competes with the ambitions of both OpenAI and Thinking Machines Lab.
The failed poaching attempt also raises questions about the sustainability of using extreme financial incentives to acquire AI talent in an era where researchers increasingly value ideological alignment and long-term impact. While Meta’s offers represented unprecedented sums—particularly Tulloch’s reported $1.5 billion package—they failed to override the sense of purpose that motivated many of Thinking Machines Lab’s early employees. This dynamic mirrors broader patterns in the tech industry, where mission-driven organizations like OpenAI and Anthropic have successfully attracted talent by emphasizing safety, transparency, and scientific rigor over short-term financial gain.
As of April 2026, Thinking Machines Lab continues to operate under Murati’s leadership, though it has lost a majority of its original founding team to Meta, OpenAI, and Elon Musk’s xAI. The company’s future valuation and ability to retain remaining staff will depend heavily on its ability to maintain a compelling research mission amid fierce competition for elite AI researchers. Meanwhile, Meta’s Superintelligence Labs faces pressure to deliver tangible results from its high-cost talent acquisition strategy, particularly as it seeks to close the gap with leaders in generative and reasoning-based AI systems.
For readers seeking to understand the evolving dynamics of AI talent competition, this case illustrates that while financial resources remain a powerful tool in recruitment, they are not determinative. The decisions made by individuals like Zoph, Metz, and Schoenholz to return to OpenAI—or by others to stay with Murati’s venture—signal that mission, culture, and intellectual autonomy continue to weigh heavily in career choices at the highest levels of AI research. As the industry matures, organizations that can balance substantial investment with authentic purpose may be best positioned to win the long-term battle for top-tier talent.
Official updates on Thinking Machines Lab’s funding rounds or Meta’s AI restructuring efforts can be found through regulatory filings with the SEC and public statements from the companies involved. As of now, no further personnel changes have been announced regarding the founding team of Thinking Machines Lab, and Meta has not disclosed additional details about the structure or goals of Superintelligence Labs beyond the launch of Muse Spark.
If you found this analysis helpful, consider sharing it with others interested in the intersection of AI, talent strategy, and corporate innovation. Comments and perspectives on how mission-driven work shapes decisions in high-stakes tech environments are welcome below.