Sam Altman and other tech leaders have warned that artificial intelligence could create a permanently unemployed "idle class," but labor economists studying real-world data say that vision is both overblown and deeply classist. Instead of fearing mass job loss from AI, experts argue the U.S. should focus on fixing its broken unemployment insurance and healthcare systems to handle whatever disruption does occur.

What Are Economists Actually Seeing in the AI Jobs Data?

Despite months of headlines about AI-driven layoffs at Amazon, Salesforce, and other major companies, the economic evidence remains murky. Labor economist Kathryn Anne Edwards explained that attributing job losses directly to AI is nearly impossible, even after the fact. "Technology in the workplace changes the nature of labor, but throughout the history of technological adoption, it is very hard to see in real time either the addition or deletion of workers," Edwards noted.

The challenge runs deeper than measurement. When companies announce layoffs and cite AI as a reason, multiple factors are usually at play simultaneously. A company might be optimistic about AI's potential and decide to operate with a leaner workforce, while also having over-hired during the pandemic boom and facing broader economic uncertainty. "All of those can be true. And then in the press release you say, 'We've adopted AI,' because that has the largest [reward] in the stock market for your shareholders," Edwards explained.

This mirrors what economists call the Solow Paradox, named after economist Robert Solow's 1980s observation that computers were everywhere in the economy yet invisible in productivity statistics. The same puzzle applies to AI today: the technology is ubiquitous in corporate announcements, yet measurable economic impact remains elusive.

Is the "AI Jobs Apocalypse" Just Hype?

Edwards and other experts reject the Silicon Valley narrative that AI will render large swaths of the workforce permanently unemployed. She was particularly critical of how tech leaders frame this possibility, calling the vision "deeply, deeply classist." The argument, she suggested, rests on the assumption that some workers are inherently more valuable or harder to replace than others, which implicitly devalues everyone else.

"I am very firmly of the belief that the U.S. worker is an incredible being, and they don't deserve to be written off by their former employer as never being able to work again. You don't talk about American workers that way in front of me, and you don't really have a place to," Edwards stated.

Kathryn Anne Edwards, Labor Economist and Independent Policy Consultant

Edwards pointed out that tech leaders often oversell their own technology's transformative power. "It feels like some of that comes from being a little high on your own supply of how amazing your technology is when you hawk it," she said. She also noted that despite all the talk about AI's revolutionary potential, there is little tangible evidence yet of measurable productivity gains flowing through the broader economy.

What Should Policymakers Actually Be Preparing For?

Rather than debating whether AI will destroy jobs, Edwards argues the real question is whether America is prepared to handle mass unemployment if it does occur. The answer, she contends, is no. The U.S. already has millions of people out of work whom the government has chosen not to help adequately, and the safety net is inadequate for a potential AI-driven disruption.

Edwards outlined the policy tools that could address this challenge:

• Unemployment Insurance Reform: Overhaul the current system to provide more robust income support and longer benefit periods for displaced workers.
• Healthcare Decoupling: Separate healthcare from employment so that job loss does not immediately result in loss of medical coverage.
• Relocation Assistance: Provide government subsidies to help out-of-work people move to regions with better job opportunities.
• Estate Tax Increases: Raise revenue through tax policy to fund these expanded safety net programs.

The irony, Edwards emphasized, is that we already know what works. "We already know what tools would work," she said. "Whether any of it happens is a question of political will".

Edwards

How to Think About AI's Real Economic Risk

Rather than fixating on whether AI will eliminate jobs, Edwards suggests reframing the conversation around preparedness. The key insight is that the magnitude of job loss matters far less than whether society has adequate systems in place to support displaced workers. "If your concern is that there's going to be job loss, and therefore we need to have a better system for handling what could be cascading layoffs, do you need to know the exact number of people who lost their job?" she asked.

Edwards

This perspective shifts the debate from speculation about AI's capabilities to concrete policy action. Edwards describes herself as an optimist, but a cynical one. She finds hope in the fact that the U.S. has not yet seriously tried to improve its safety net. "The lowest bar is the greatest source of optimism," she explained, "because I'd feel very differently if we had tried anything".

Edwards

The contrast between tech leaders' warnings and economists' measured skepticism reflects a fundamental disagreement about what the real problem is. Altman and others warn of technological displacement; Edwards warns of policy failure. Both could be true, but only one is within our immediate control.