Zuckerberg’s $1.5 Billion Offer Rejected by Mira Murati’s Top AI Engineer

You should never bet against Mark Zuckerberg, or so runs a popular saying in Silicon Valley. He’s known for his relentless drive and usually finds a way to get what he wants. But this time, things didn’t go as planned, and reports say he’s not happy about it. Zuckerberg recently approached former OpenAI CTO Mira Murati with an offer to buy her AI startup, Thinking Machines Lab. When she turned him down, he switched tactics and tried to hire her top engineer instead with a jaw-dropping offer of $1.5 billion.

Zuckerberg’s offer to acquire Murati’s startup was around USD 1 billion, according to US media. After she declined, Meta reportedly began reaching out to several employees of the company, trying to attract them with huge salaries and bonuses. One of the key targets in this recruitment effort was Andrew Tulloch, a cofounder of Thinking Machines and a well-known expert in artificial intelligence.

Massive Offer, But Tulloch Said No

Tulloch, who helped build Thinking Machines Lab with Murati, is recognised in the AI field for his work on machine learning systems and his role in creating PyTorch, a major tool used in AI research. Interestingly, Tulloch had previously worked at Meta and left the company before cofounding his current venture. Meta’s offer to him reportedly included compensation worth USD 1.5 billion over six years.

Despite the enormous number, Tulloch turned the offer down. According to reports, Zuckerberg himself had reached out to him, hoping to bring him back to Meta as part of the company’s new AI push. Even Alexandr Wang, who now heads Meta’s Superintelligence Lab, attempted to convince Tulloch to return. However, Tulloch wasn’t interested, likely because of his commitment to his new company and possibly his ownership stake in Thinking Machines Lab, which is already valued at more than $30 billion.

Tulloch’s background is impressive. He graduated with top honours in mathematics from the University of Sydney and holds a master’s degree from Cambridge. He is also working toward a Ph.D. at UC Berkeley. His time at Meta, from 2012 to 2023, gave him deep experience in building AI tools. He also spent a year at OpenAI, contributing to GPT-4’s development, before joining forces with Murati to start Thinking Machines Lab in early 2025.

Meta’s AI Talent Hunt Goes Beyond One Company

Meta’s attempts to hire Tulloch are part of a larger strategy. The company is trying to build a “Superintelligence” team by hiring some of the best AI experts from across the industry. It has reportedly reached out to over 100 OpenAI staff and has successfully hired at least 10 of them. Other startups, including Dario Amodei’s Anthropic, have also seen Meta trying to attract their engineers. Meta spokesperson Andy Stone denied the claims about trying to buy Murati’s startup. He also called the $1.5 billion figure for Tulloch’s offer “inaccurate and ridiculous”, saying that any compensation offers depend on company stock performance.

Not Everyone Said No

While Tulloch refused the offer, others eventually accepted Meta’s proposals. One such person was 24-year-old Matt Deitke. Initially, Deitke turned down Meta’s offer of USD 125 million. But after a personal approach from Zuckerberg, who promised to double the deal, Deitke agreed. The final offer reportedly reached USD 250 million, with USD 100 million guaranteed in the first year. Deitke’s decision to join Meta shows just how serious and personal Zuckerberg is about building his next generation of AI talent. As the AI race heats up, Meta’s aggressive recruitment strategies show the intense competition among tech giants to secure the best minds in artificial intelligence. With billions on the line, this battle for talent is far from over.