In listing after listing, Travis Hart kept seeing the same job requirement: Applicants should know how to use ChatGPT.

After losing his job at a small e-commerce company where the growing reliance on artificial intelligence concerned him, he was seeing the technology crop up throughout his employment search.

“It was scary going through all these jobs, and people are listing it as a skill,” said Hart, 34, of Monticello, Minn. “There’s a trick to it, but it’s being observed as a skill, which — it’s not the same thing.”

Skill or not, generative AI platforms like ChatGPT are rapidly infiltrating the American workplace. Companies — from health care to financial services to communications — are investing billions of dollars as they adopt new technology, fill AI-related jobs and retrain current employees.

Retail behemoth Walmart plans to spend nearly $1 billion on training in 2026, and has announced a partnership with OpenAI. Bank of America is spending $4 billion, or about a third of its technology budget, on AI this year.

The rapid growth of AI is putting worker anxiety — already elevated thanks to a slowing labor market— on overdrive.

The technology’s development has echoes of the automation and offshoring boom of the late 20th century, promising increased efficiencies at the cost of human jobs. The latter happened over the span of decades and largely affected blue-collar workers. AI is evolving much faster, and its disruption extends to white-collar jobs.

Some business leaders have thrown lighter fluid on the discussion, predicting the technology will replace many desk jobs altogether.

At the Aspen Ideas Festival this summer, Ford Motors CEO Jim Farley said AI “is going to replace literally half of all white-collar workers in the U.S.” A September Resume.org survey of U.S. business leaders found four in 10 companies plan to replace workers with AI by 2026.

Robots and recessions

Though employers have been hesitant to hire amid widespread economic uncertainty this year, they are investing in AI — “one of the few areas propping up the economy” as tariff shocks hit, according to a research note from Oxford Economics senior economist Makoto Tsuchiya.

Neel Kashkari, president of the Federal Reserve Bank of Minneapolis, described the trend in a recent essay as reallocation “away from labor-intensive industries to less labor-intensive industries.”

“We have seen a remarkable investment boom in technology — building data centers for AI, for example," Kashkari wrote. “While it takes a lot of people to build a new data center, it takes relatively few to operate one.”

The mass layoffs that accompany an economic downturn haven’t happened, but a growing number of jobseekers have spent more than six months looking for work. As hiring slips, the national unemployment rate is ticking up.

If the country tips into a recession, AI could upend the trajectory of a typical recovery.

“It’s not obvious that (laid off workers) will get their job back,” said Stefania Albanesi, an economics professor at the University of Miami.

As businesses ramp up after a downturn, she said, it’s typical for employers to invest in technology to replace the workers they let go.

“If we have a recession where firms lay off a bunch of workers, and then in three years, this technology has improved a lot, that’s going to be a little bit of a catalyst for a deployment more at scale of that technology,” she said.

Part of the challenge for workers and employers, like the economy as a whole, is that it’s not clear where AI will ultimately land. As use has exploded, so has the technology’s capabilities, doing everything from answering a question or writing an email to building a slide deck or reading a resume.

More than a third of respondents to a new Deloitte consumer survey said they use the technology for work, up from 6% in 2023, with many incorporating the tools on their own.

Though nearly 70% of the 3,500 survey respondents agreed tech companies are innovating too quickly and not doing enough to mitigate risks, most agreed “they can see gen AI serving as a valuable assistant in their future roles.”

“The pace of this is is dramatically different than other experiences we’ve collectively had,” said Lou DiLorenzo, who leads the U.S. tech, AI and data strategy practice at Deloitte. “The contribution (gen AI) can make to solving different problems is evolving rapidly and continues to show no signs of slowing down.”

Quick change

Minneapolis-based Vilas AI started conducting workplace AI training in 2023, when the technology was “really awkward,” said co-owner James Holmberg. “Now we’re in 2025 — it took such a huge jump,” he said.

Still, Holmberg said, he hasn’t seen AI replace jobs, and he doesn’t necessarily expect that.

“I don’t think these things go away,” he said. “I think [jobs] actually get leaner, you know? They just get more intense.”

As a software engineer, 33-year-old Daniel Martelly of Minneapolis knows a bit more about AI than the average user. He recalled grappling with what the technology might mean for his work, from skepticism to fear to acceptance.

“I tried it out, and I was like, OK, I can see some really cool things that this is doing,” Martelly said. “And now I have gone through the whole emotional cycle, and I can look at it a little bit more objectively.”

While there’s lingering worry that AI could replace him — “If my boss can do it instead of me doing it, what happens to me?” he said — the technology’s limitations are clear. He’s more concerned about younger colleagues turning to AI instead of learning the fundamentals.

“It’s important to understand what’s happening to truly not just get it working, but understand why it’s working,“ Martelly said. ”And then what that means is, when it’s broken, you can understand why it’s broken and really get to the correct answer."

_____

©2025 The Minnesota Star Tribune. Visit at startribune.com. Distributed by Tribune Content Agency, LLC.