Americans worry AI is coming for these jobs

The boom of artificial intelligence has ignited a wave of new possibilities, but it’s also stirring fears among many Americans about their jobs. One of the main concerns is that AI could put humans out of work, and new data shows which jobs people worry about the most.

So who’s likely at risk?

Seventy-three percent of U.S. adults think that AI will probably lead to fewer jobs for cashiers in the next 20 years. Fifty-nine percent say journalists, 67% say factory workers, and 48% say software engineers, according to a survey from the Pew Research Center. AI experts who were also surveyed mostly agree with the public on those positions, though there are some jobs where they differ. Sixty-two percent of AI experts, vs. 33% of the public believe AI will result in job losses for truck drivers.

But Molly Kinder, a fellow at the Brookings Institution who has been examining the impact of generative AI on work, says some of these opinions are conflating old automation with what generative AI can do.

“As a country, our narrative about AI and work is still centered on yesterday’s automation,” Kinder said. “Generative AI is not your grandparents’ automation. It’s moving up the skill chain … in ways I don’t think the public is fully aware of.”

The difference boils down to the fact that old automation heavily hit blue collar jobs — think machines that could accelerate manufacturing. But generative AI is going to have a big impact on white collar work because it’s able to summarize complex and lengthy documents, crunch large sets of data, and interact conversationally.

Kinder worked with colleagues to forecast which jobs are most exposed to generative AI based on the tasks the technology does best. Their research suggests the most exposed jobs are in office and administrative support, legal, sales, computers and math, and business and financial sectors. That could include jobs such as office clerks, travel agents, paralegals, computer programmers, and tax preparers.

Mark Muro, a Brookings fellow and co-author of the report, said the findings suggest only that these jobs will be most exposed to AI, but it’s unclear whether it will aid workers or replace them.

“It’s these white collar, process-oriented jobs that are a large part of America’s middle class that at least need to be thinking about this.”

It’s difficult to predict just how vast the impact will be as AI capabilities continue to expand, said Eric So, a professor of global economics and management at the Massachusetts Institute of Technology who studies how human nature and technology interact. Already, in just a handful of years, generative AI has gone from a text chatbot to being able to produce realistic images, understand and respond via voice, and even control a mouse or keyboard to interact with what’s on screen. With the rise of agentic AI, or bots that can function autonomously without human intervention, there may be even more surprises coming down the road. The hope is that it won’t be ordering us overpriced eggs without our knowledge.

“Once you get a sense of those capabilities and where they’re heading … this is truly a profound societal change,” he said.

That said, adoption of generative AI has been slower than many expected. While some jobs may be seeing the impact already (evidence suggests that AI might be eliminating various computer programming roles), others are taking measured, conservative approaches, researchers say. That’s especially true in fields where there may be more regulation like medicine and law.

Still, if one thing has remained consistent, it’s that people’s anxiety about the technology is rising. About 4 in 10 people said they were concerned about the increased usage of AI back in 2021 and 2022, Pew’s surveys show. Today, that number has jumped to 51%. And much of that fear revolves around jobs.

“This theme of public anxiety around jobs, job loss, and the way people work has been present going back several years,” said Colleen McClain, a Pew senior researcher.

The concerns are justified, Muro said, especially because the technology could raise the bar for some white collar entry-level jobs or even change the kinds of jobs that are open to future generations. But don’t panic — there will always be a role for humans to play even if we don’t necessarily know what those will be yet, Kinder said.

“Part of the reason there’s so much anxiety is the pace of the innovation and how much experts can’t even predict,” she said. But “the more people actually touch these technologies ... the more they can see the possibilities.”