日本語
한국어
Tiếng Việt
简体中文
The intensifying talent war in Silicon Valley’s artificial intelligence sector has redefined compensation standards, making even the most significant scientific achievements of the past seem financially modest. Recent reports highlight a staggering offer from Meta to AI researcher Matt Deitke: an estimated $250 million over four years, potentially including $100 million in the first year alone. This unprecedented package shatters all previous records for scientific and technical remuneration, including earnings during the foundational scientific endeavors of the 20th century.
Deitke, who co-founded the startup Vercept and previously spearheaded the multimodal AI system Molmo at the Allen Institute for Artificial Intelligence, possesses highly sought-after expertise in systems that integrate images, sounds, and text – precisely the kind of advanced technology tech giants are racing to develop. His case is not isolated; Meta CEO Mark Zuckerberg reportedly extended an offer of $1 billion in compensation over several years to another unnamed AI engineer. This raises a crucial question: What drives such extraordinary valuations?
The Trillion-Dollar Race for Artificial General Intelligence (AGI)
These astronomical sums reflect the high stakes perceived by leading tech companies. They are engaged in an intense global race to achieve artificial general intelligence (AGI) or superintelligence – machines capable of performing intellectual tasks at or beyond human levels. Companies like Meta, Google, and OpenAI are convinced that whoever reaches this technological breakthrough first stands to dominate markets potentially worth trillions. Whether this vision is a realistic projection or a product of Silicon Valley hype, it undeniably fuels an unparalleled surge in AI talent compensation.
To contextualize these modern AI salaries: J. Robert Oppenheimer, the scientific director of the Manhattan Project, earned approximately $10,000 annually in 1943. When adjusted for inflation, this amounts to roughly $190,865 in today’s dollars – a figure comparable to what a senior software engineer might earn now. In stark contrast, the 24-year-old Deitke, a recent PhD dropout, is set to earn approximately 327 times Oppenheimer’s salary during the development of the atomic bomb.
AI Talent Outearns Pro Athletes and Tech Pioneers
The compensation for top AI researchers also eclipses that of many highly paid professional athletes. For instance, Steph Curry’s most recent four-year contract with the Golden State Warriors was reportedly $35 million less than Deitke’s Meta deal, leading observers to describe the AI talent market as an “NBA-style” competition, where AI experts often command more than sports stars.
Mark Zuckerberg has explicitly stated Meta’s commitment to investing heavily in AI talent, driven by the conviction that “superintelligence is going to improve every aspect of what we do.” This strategic focus explains why AI researchers are viewed not merely as well-compensated professionals, but as irreplaceable assets. If these companies are correct, the first to achieve AGI or superintelligence will possess technology capable of creating endless new products, automating millions of knowledge-worker jobs, and fundamentally transforming the global economy. Such a company could become the wealthiest in history.
Even the peak salaries of legendary figures from the early tech era pale in comparison. Thomas Watson Sr., the iconic CEO of IBM, received $517,221 in 1941, the third-highest salary in America at the time (approximately $11.8 million in 2025 dollars). Today’s leading AI researcher packages can exceed five times Watson’s top earnings, despite his pivotal role in building one of the 20th century’s most dominant technology companies.
The collaborative innovation of eras like Bell Labs’ golden age, which yielded the transistor and information theory, also saw different compensation structures. Claude Shannon, who laid the mathematical foundation for modern communication with information theory in 1948, worked on a standard professional salary. Furthermore, the “Traitorous Eight” who founded Fairchild Semiconductor, essentially birthing Silicon Valley, began with a seed funding of $1.38 million (around $16.1 million today) for the entire company – a sum now dwarfed by the compensation of a single AI researcher.
Space Race Salaries: A Different Universe of Pay
The Apollo program offers another striking salary comparison. Neil Armstrong, the first person to walk on the moon, earned approximately $27,000 annually, which translates to about $244,639 in today’s money. His fellow crew members, Buzz Aldrin and Michael Collins, earned even less. While current NASA astronauts earn between $104,898 and $161,141 annually, a top AI researcher at Meta could earn more in just three days than Armstrong made in a year for his historic “one giant leap for mankind.”
Engineers behind the Apollo program’s rockets and mission control systems also received modest salaries by contemporary standards. A 1970 NASA technical report indicates that a newly graduated engineer in 1966 started with $8,500 to $10,000 annually (roughly $84,622 to $99,555 today). Even the most elite, experienced engineers with two decades of service peaked at around $278,000 per year in today’s dollars – an amount a leading AI researcher can now earn in a matter of days.
Understanding the Unique Dynamics of the AI Talent Market
While premium compensation for technical talent isn’t entirely new – for instance, in 2012, three University of Toronto academics were “auctioned” to Google for $44 million (approximately $62.6 million today) – the current figures are unprecedented. Several factors converge to explain this explosion in AI salaries:
- Unparalleled Wealth Concentration: The tech sector operates with an industrial wealth concentration not seen since the Gilded Age, allowing companies with trillion-dollar valuations to engage in intense bidding wars.
- Extreme Talent Scarcity: The pool of researchers with the highly specialized expertise needed for cutting-edge AI systems, particularly in areas like multimodal AI, is exceptionally limited.
- Exaggerated Market Hype: AI is currently viewed as the “next big thing” in technology, driving speculative investment and inflated valuations.
- Strategic Investment: Unlike past projects with defined costs, such as the Manhattan Project ($1.9 billion total, or $34.4 billion adjusted for inflation), companies like Meta plan to spend tens of billions annually on AI infrastructure. For a company nearing a $2 trillion market cap, the potential returns from achieving AGI first far outweigh individual compensation packages. As one executive put it, “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.”
- “Arms Race” Mentality: Tech giants perceive themselves to be in an existential arms race, where the victor could fundamentally reshape civilization. Unlike the finite goals of the Manhattan Project or Apollo program, the pursuit of AGI ostensibly has no ceiling. A machine achieving human-level intelligence could theoretically improve itself exponentially, potentially leading to an “intelligence explosion” and cascading discoveries.
Young researchers actively leverage this market dynamic, sharing offer details and negotiation strategies in private online groups and even hiring unofficial agents. Companies sweeten deals not only with massive cash and stock but also with immense computing resources, such as promises of access to 30,000 GPUs – the specialized chips crucial for AI development.
Whether these companies are on the cusp of creating humanity’s ultimate labor replacement technology or merely chasing an elaborate hype cycle remains an open question. However, the journey from Neil Armstrong’s modest $8 per diem (about $70.51 today) during his moon mission to today’s multi-million-dollar AI salaries marks a profound shift in how groundbreaking scientific and technical talent is valued. As Vercept co-founder Kiana Ehsani quipped after Deitke accepted Meta’s offer, “We look forward to joining Matt on his private island next year.”