Take that, Oppenheimer: Meta offers AI researcher $250M over 4 years in talent war

Meta recently offered AI researcher Matt Deitke $250 million over four years—an average of $62.5 million annually—shattering every historical precedent for scientific compensation. The 24-year-old’s package is 327 times what Manhattan Project leader J. Robert Oppenheimer earned while developing the atomic bomb, reflecting Silicon Valley’s belief that the race for artificial general intelligence could reshape civilization and create trillions in market value.

The big picture: Tech companies are treating AI talent like irreplaceable assets rather than well-compensated professionals, driven by the conviction that whoever achieves artificial general intelligence first could dominate markets worth trillions.

• Meta CEO Mark Zuckerberg reportedly offered another unnamed AI engineer $1 billion in compensation spread over several years.
• Companies believe they’re in an arms race where the winner could control technology capable of automating millions of knowledge-worker jobs and transforming the global economy.
• Zuckerberg told investors Meta will continue “throwing money at AI talent because we have conviction that superintelligence is going to improve every aspect of what we do.”

Historical context shows unprecedented compensation levels: Past scientific achievements earned modest salaries compared to today’s AI packages.

• J. Robert Oppenheimer earned approximately $10,000 per year in 1943 ($190,865 in today’s dollars) while leading the Manhattan Project that developed the atomic bomb.
• Neil Armstrong made about $27,000 annually ($244,639 today) as the first person to walk on the moon.
• Apollo program engineers with 20 years of experience peaked at around $278,000 per year in today’s dollars—what Deitke can now earn in just a few days.

What makes this talent market different: Several factors explain the unprecedented compensation explosion beyond typical tech premiums.

• Only a small number of researchers have the specific expertise needed for the most capable AI systems, particularly in multimodal AI that combines images, sounds, and text.
• Multiple companies with trillion-dollar valuations are competing for an extremely limited talent pool.
• The economics differ fundamentally—while the Manhattan Project cost $1.9 billion total, Meta alone plans to spend tens of billions annually on AI infrastructure.

Who’s Matt Deitke: The 24-year-old recently dropped out of a PhD program and cofounded a startup called Vercept.

• He previously led development of Molmo, a multimodal AI system, at the Allen Institute for Artificial Intelligence, a research organization focused on AI development.
• His expertise in systems that handle images, sounds, and text made him a prime target for Meta’s recruitment efforts.
• His compensation package is $35 million more than NBA star Steph Curry’s most recent four-year contract with the Golden State Warriors.

How the talent war operates: Young researchers have developed sophisticated networks and negotiation strategies to maximize their value.

• They maintain private chat groups on Slack and Discord to share offer details and negotiation strategies.
• Some hire unofficial agents to represent them in negotiations.
• Companies offer not just massive cash and stock packages but also computing resources—some potential hires were promised access to 30,000 GPUs, the specialized chips that power AI development.

What they’re saying: Industry executives justify the astronomical spending as necessary investment in potential civilization-changing technology.

• One executive told The New York Times: “If I’m Zuck and I’m spending $80 billion in one year on capital expenditures alone, is it worth kicking in another $5 billion or more to acquire a truly world-class team to bring the company to the next level? The answer is obviously yes.”
• After Deitke accepted Meta’s offer, Vercept co-founder Kiana Ehsani joked on social media: “We look forward to joining Matt on his private island next year.”