The Nvidia chief says electricians, welders and builders are the first big winners of the AI boom. The construction data backs the direction of his claim. Whether those jobs last is a different question.
Nvidia CEO Jensen Huang says the artificial intelligence boom is already putting skilled tradespeople to work in large numbers, a side of the AI jobs story he argues most people are missing.
In a video clip circulating on X, Huang says the purpose of work has not changed, only the tools. He points to the wave of chip plants, computer factories and data centers needed to run AI. “Electricians, construction workers, welders, technicians, networking engineers, all of these jobs are now in huge demand,” he says. “We’ve created about half a million of them. We’re probably gonna create a lot more.”
It is a message Jensen Huang has been repeating all year, and the figures he uses have grown as he goes.
What Jensen Huang has said in 2026
January, Davos. Speaking with BlackRock CEO Larry Fink at the World Economic Forum, Huang called the AI buildout the “largest infrastructure build-out in human history” and named plumbers, electricians, construction workers and steelworkers as the people who would build it. “Everybody should be able to make a great living,” he said. “You don’t need to have a PhD in computer science to do so.”
Spring. Huang said AI had “created more than half a million jobs in the last couple of years,” according to The National. In May he told Carnegie Mellon graduates that it was the moment for “electricians, plumbers, iron workers, technicians, builders.”
August. On Fox Business, Huang said “hundreds of thousands of jobs are being created as we speak” in chip plants, packaging and AI factories, while conceding that “some jobs will be eliminated.”
September, the White House. Sitting beside Elon Musk at a gathering of tech executives on September 29, Jensen Huang raised his estimate, saying data center construction could create “probably about a million jobs,” many of them blue-collar. “This is the first time in probably 50 years we’re re-industrializing the United States,” he said.
Do the numbers back Jensen Huang up?
On direction, yes. Construction starts for U.S. data centers reached $84.1 billion in the first seven months of 2026, nearly three times the level of a year earlier, according to ConstructConnect. Big Tech’s spending on data centers and related infrastructure could approach $700 billion this year.
The labor squeeze is real too. Industry projections estimate that 2.1 million skilled trade jobs could go unfilled in the United States by 2030, including roughly 130,000 additional electrical workers. Data center builder QTS, owned by Blackstone, had about 10,000 workers on its job sites a year ago and expects 40,000 by the end of 2026.
Tech companies are now paying to train the workforce they need. In June, Google committed $50 million through Google.org to train more than 300,000 electricians, pipefitters, sheet metal workers and other tradespeople across more than 20 states, working with 14 unions and four trade associations.
What the data does not do is confirm Jensen Huang’s exact totals. Neither the half-million nor the million figure comes from a published count, and Nvidia has not released the method behind them. Treat them as his estimates.
The catch in Jensen Huang’s numbers: building is not running
Much of this hiring happens during construction. Once a data center is built, it runs with a far smaller staff, and the Brookings Institution has warned that tradespeople on these projects have no guarantee of permanent work when the build ends. Data center construction also slowed last year for the first time since 2020, according to Bloomberg, as developers ran into zoning, permit and power delays.
There is also the question of who is talking. Nvidia sells the chips that fill these buildings, and its growth depends on companies continuing to spend heavily on them. A Jensen Huang forecast that more data centers mean more good jobs is also a forecast that more data centers mean more Nvidia sales.
Not every tech leader shares his optimism. Bill Gates has warned that AI could drive large-scale unemployment, especially in office work. Huang’s answer is that history is on his side: productivity gains have tended to create more jobs than they destroy, and the first, most visible effect of AI on the labor market is people building things. It is part of a wider pitch from tech leaders, including Jeff Bezos’s prediction of a three-day workweek, that AI will be good for workers.
What Jensen Huang’s claim means for workers
For young people weighing college against an apprenticeship, the near-term signal from Jensen Huang and the hiring data is clear: electricians, welders and other trades are in demand, and the companies building AI are competing for them. The longer-term picture, including how many of those jobs remain once the construction wave passes and what happens to office jobs that AI can do, is still being written.
Sources: Video clip shared on X by @vikktorrrre; Fortune, January 2026; Let’s Data Science, citing The National; Yahoo Finance; Fortune, September 2026; The Wealth Advisor; ConstructConnect; The Next Web.
Leave a Reply