100 million-dollar pay packages aren't just for the Shohei Ohtanis and Cristiano Ronaldos of the world anymore. In Silicon Valley, nine-figure pay days are reportedly now being floated to the world's top talent as the race to own AI enters a new frenzied stage.
Meta has made at least 10 high-pay offers of up to $300 million over four years to top OpenAI researchers for what it's calling its Superintelligence Lab, Wired reported this week. Sam Altman claimed in June that OpenAI workers had been offered $100 million signing bonuses to jump ship. Meta spokesman Andy Stone called news of the reported pay "untrue," saying "the size and structure of these compensation packages have been misrepresented all over the place."
Whatever the actual figures are, it seems a select few researchers could see bank account balances that rival or surpass CEOs at other Big Tech companies — and they would out-earn other tech workers by numbers that are hard to envision. (A BI analysis of Meta's federal filings last week found that software engineers at the company can make up to $480,000 in base salary.) A former Meta engineer who still works in the tech industry told us that as "top tech talent is finally being treated like top sports" many of tech's rank and file workers are filled with resentment and jealousy, "especially amongst folks that have been in the industry a while."
Demand for AI experts far outpaces the supply. Whatever enticing salaries Meta is actually offering must be outsized enough that they tempt people to abandon the sleek, hotbed of AI innovation to take on the task of bringing Meta, a company that bet everything on the metaverse, through yet another rebrand. To achieve Mark Zuckerberg's dream of building a personalized AI for everyone, Meta no doubt needs to open its wallet.
Zuckerberg seems to be testing now how many zeros he needs to put on a check to get people to come play for his team. But perhaps more than anything, the eye-popping salaries underscore how the divide between top AI talent and the rest of us is becoming not just a gap, but a chasm.
"It doesn't make sense for engineers, no matter how experienced and with great credentials, to make NBA players' salaries," says Natalia Luka, who studies economic sociology, organizations, and science and technology at University of California, Berkeley. The "exorbitant sums" that tech companies are paying, "not just for AI engineers, but also the computing power, the data centers, all of those costs," she says, is putting enormous "pressure on them to cut costs elsewhere."
<es-blockquote data-quote=""There's this existential dread going around that the field is fundamentally changing," says a former Meta engineer." data-styles="pullquote-breakout" data-source="">
"There's this existential dread going around that the field is fundamentally changing," says a former Meta engineer.
</es-blockquote>
Just as top AI employees have humungous offer letters sent their way, thousands of tech employees are watching their job security crumble. Microsoft announced last week that it will lay off 9,000 workers (with the sales and Xbox divisions among those affected), bringing the total number of cuts at the company so far this year to about 15,000. These come as tech companies have spent nearly three years culling their ranks after overhiring and hoarding talent during the height of the Covid-19 pandemic. More than 600,000 tech workers have lost their jobs since Layoffs.fyi started tracking cuts in the industry in 2022. Tech jobs overall, however, have continued to grow, and are expected to do so twice as fast as other sectors over the next decade, according to a new report from CompTIA, a trade association for the IT industry.
Still, the disparity has alarmed many. "There's this existential dread going around that the field is fundamentally changing, and the new entrants have had the rug pulled from under them as most companies are only hiring senior folks," the former Meta engineer told us. While that "makes sense in the short term," the person added, they fear it "is only going to make it impossible for them to grow the next generation of senior engineers."
Some Meta insiders are deeply cynical about the new "Superintelligence" organization led by 28-year-old Alexandr Wang and recent hires from OpenAI and DeepMind. Screenshots from a group of Meta employees on Blind shared with Business Insider show one employee calling the new group "marketing BS to feed the media." Some are concerned that Meta's current GenAI org will be sidelined or laid off, with one post asking, "Should I switch away from GenAI? I feel like we're all going to get fired."
Companies everywhere have increasingly made investments and pivots to AI over the past few years, and other workers have been cut to foot the bills. Meta bought nearly half of Scale AI last month for $14 billion. Big Tech companies from Google to Meta have boasted about the ways they're using AI to write code and become more efficient, all while the career ladder for entry-level software engineers topples. "You probably have to assume that just given the intense focus on all things AI right now, that it does diminish other potential areas of innovation, other potential investments, and by extension, it probably is going to diminish other workers in some capacity," says Tim Herbert, chief research officer at CompTIA.
The concurrent battle between Meta and OpenAI over the industry's most elite talent is far from the only one — companies are willing to pay more for AI talent at much lower levels, too. As of April 2024, entry level AI engineers made about 8.5% more than other engineers, according to Levels.fyi. Mid- and senior-level engineers earn about 11% more than similarly experienced engineers not working directly on AI. Demand for AI skills has grown by 21% annually since 2019, according to management consulting firm Bain & Company.
Not everyone is mad about the pay gap. A current engineer in Meta's GenAI org told us they believe most people at the company understand the rationale, and even support it. "I cannot produce that kind of impact and hence do not deserve that kind of compensation," they said. "I think most Meta employees are pretty much on board with this. If this team delivers disproportionately, we all benefit via stocks."
The rush for AI talent isn't so different from other eras of rapid tech innovation, where few have the skills to perform highly sought after work, says Sonny Tambe, a professor at the Wharton School of the University of Pennsylvania. "What's different now is that the pace is much faster and there could be outsized rewards for companies who win this market, so the stakes are unusually high," Tambe says. "AI companies don't have time to wait for the talent pool to expand, and so the effect is more pronounced."
Experts are skeptical that superstar athlete salaries will become the new norm for top AI talent; as more people are trained to lead generative AI teams, this may even become a passing fad. "The market is going to adjust in terms of having credentialed people who can do this kind of work," Luka says. "Right now, it really is a fairly select group that knows how to run these giant AI systems."
While we wait for more AI experts to emerge, there seems to be a two-tier system splitting inside Meta. Workers on Blind are describing Superintelligence recruits as "the chosen few." One employee sarcastically noted that those working in the GenAI org will "get to label data for minimum wage." Another noted, "Only a select few will get promoted… this is the era of elite internal poaching." Tech companies will still need to invest in other areas to move their goals forward. For now, it's a great time to be one of the few brightest brains on AI.
Amanda Hoover is a senior correspondent at Business Insider covering the tech industry. She writes about the biggest tech companies and trends.
Pranav Dixit is the Meta Correspondent at Business Insider based in the San Francisco Bay Area.
The post Welcome to Silicon Valley's winner-take-all era appeared first on Business Insider