Across Silicon Valley, a hidden and costly war is underway, fueled by a secret “list” of the brightest artificial intelligence engineers and researchers. Mark Zuckerberg and his rivals are ready to pay millions, even hundreds of millions, to recruit them.
Article Contents:
- What is the “list” shaking Silicon Valley?
- Who are these geniuses?
- Zuckerberg’s fierce chase for talent
- A secretive world of research and security precautions
- Why are they worth hundreds of millions?
- What motivates these researchers? It’s not just money
- “Tribal knowledge”: A small and tight-knit community
What is the “list” shaking Silicon Valley?
Across Silicon Valley, the brightest minds in artificial intelligence are talking about «the list», a compilation of the most talented engineers and researchers in the field that Meta CEO Mark Zuckerberg has spent months putting together. This list is not just a roster of names; it is a treasure map in the frantic race toward supremacy in the AI revolution.
Everyone wants to hire them, from tech giants to venture capitalists. But no one is pursuing them with more enthusiasm than Zuckerberg, who is trying to poach them from the world’s top research labs, offering compensation packages reaching $100 million for a select few prominent stars.
Who are these geniuses?
Recruits on “the list” typically hold PhDs from elite universities like Berkeley and Carnegie Mellon. They have experience at places like OpenAI in San Francisco and Google DeepMind in London. They are frequently in their twenties and thirties, and they know each other well. Names like Lucas Beyer, who describes himself as a “scientist dedicated to creating masterpieces,” and Misha Belinko, an expert in large-scale machine learning, are hard currency in this market.
These researchers spend their days solving complex problems that require massive computational power, and their talents—which were once obscure—have never been as valuable as they are today.
Zuckerberg’s fierce chase for talent
Zuckerberg aspires to bring these geniuses into Meta’s new lab focusing on “superintelligence,” an artificial intelligence smarter than humans. This frantic pursuit comes after the launch of Meta’s latest large language model, Llama 4, proved disappointing. One recruit who spoke with Zuckerberg personally described the company’s goal as nothing less than “a blood transfusion from the best AI labs in the country.”}
Zuckerberg personally reaches out to his dream team of potential candidates, using a chat group with two Meta executives called “Recruiting Party 🎉” to discuss hundreds of candidates and outreach tactics via email or text message.
A secretive world of research and security precautions
In this secretive world of building powerful AI models, labs have begun taking drastic steps to protect their scientific breakthroughs. At Anthropic and OpenAI, researchers work on separate, restricted-access floors, often with curtains drawn to protect against prying eyes. Anthropic leaders have grown so worried that their researchers might be targets for foreign espionage that they once invited an FBI agent to brief employees on the risks.
Why are they worth hundreds of millions?
The pool of engineers with deep expertise in this type of AI research is very small. One reason companies are willing to pay these astronomical sums is that the cost of an elite team of AI engineers represents only a fraction of the price of AI infrastructure like data centers. Meta alone plans to invest roughly $70 billion this year in artificial intelligence, while Amazon, Microsoft, and Alphabet are spending even more. Alongside hardware costs, the human element looks like a bargain.
It is difficult to gauge the value of these researchers, whose work most people know nothing about. But executives can search Google Scholar, browse their research papers, and check how many times their work has been cited.
What motivates these researchers? It’s not just money
People in the interconnected AI community say researchers—whose skills have never been this lucrative—are not primarily driven by money. Until recently, many were perfectly happy being university professors. Today, they are heading to AI labs not just for the paycheck, but because tech companies with limitless resources are the only ones providing the computing power, data, infrastructure, and freedom they need to run experiments and scale their models.
Alexei Efros, a professor at UC Berkeley, says: “For my students, the goal was never to do trendy things and become a millionaire. The goal is trying to solve cool, important, and unsolved problems.”
“Tribal knowledge”: A small and tight-knit community
These researchers’ loyalty to one another transcends companies. When deciding to leave for Meta, they compare notes, exchange information, and plan their futures together. A small group of the smartest AI researchers has built what one described as “tribal knowledge” that is nearly impossible to replicate. They have lived in the same communal houses in San Francisco, discussing research papers that might provide clues to achieving the next great breakthrough.
Despite Zuckerberg’s efforts, he has not had major success hiring top field stars, such as OpenAI co-founder Ilya Sutskever. However, he recently succeeded in hiring three researchers from OpenAI’s Zurich office: Lucas Beyer, Alexander Kolesnikov, and Xiaohua Zhai. With every failed attempt, Zuckerberg simply moves on to the next person on “the list.”}
