White-Collar Jobs Face First Wave of AI Displacement, Brookings Researcher Warns

The coming wave of AI-driven job displacement will strike hardest at the workers who felt most secure during previous automation cycles: college-educated professionals whose work happens primarily on computers.

That's the central argument from Molly Kinder, who has led a three-year research project at the Brookings Institution examining how generative AI is transforming work. In a recent interview on the Platformer podcast, Kinder outlined what she calls the "messy middle"—a potentially decades-long transition period between today's largely intact labor market and the post-AGI abundance promised by Silicon Valley.

The inversion of pandemic-era safety

Kinder's thesis represents a striking reversal from the pandemic experience. During COVID-19, workers who could perform their jobs remotely from home were the safest, while essential workers faced the greatest exposure. Now, that calculus has flipped.

"If you can do your job locked in a closet with a computer, eventually you're probably going to be in trouble," Kinder said. She points to OpenAI's task exposure data showing that knowledge sectors—law, finance, consulting, sales—face the highest risk from large language models like ChatGPT.

Meanwhile, jobs requiring physical presence—restaurant workers, hair stylists, repair technicians—show minimal exposure to current AI capabilities. While robotics may eventually automate physical labor, Kinder sees computer-based knowledge work as the immediate frontier.

Why the American dream is at stake

The implications extend beyond individual job losses. For decades, the path to middle-class security ran through higher education and white-collar employment. Computers enhanced knowledge workers' productivity without replacing their core cognitive functions. A lawyer in the 1980s might have lost their secretary to word processing software, but the lawyer's own job became more valuable.

Large language models threaten to break that pattern by potentially substituting for specialized cognition itself. "We've called this skill-biased technological change," Kinder explained. "Up until the moment ChatGPT was launched, I always considered computers as boosting the knowledge worker. But the computer didn't do my job. My brain did my job."

This shift carries particular weight given current economic anxieties around affordability. With manufacturing and clerical jobs already hollowed out, college-educated professional roles represent one of the few remaining paths to homeownership and family financial security. Kinder has interviewed numerous young people expressing fear that AI will eliminate these opportunities despite their following every prescribed step—good grades, strategic major selection, degree completion.

Beyond universal basic income

Kinder rejects universal basic income as a solution, arguing that checks large enough to replace software engineer salaries would destroy the labor market by removing incentives for essential workers to continue policing streets, building houses, or staffing hospitals.

Instead, she advocates targeted interventions: workforce reinvestment funds requiring companies to pay for white-collar apprenticeships when cutting young workers, wage insurance for older displaced workers, and potentially public job creation programs if good positions grow genuinely scarce.

Why it matters

The concentration of AI displacement in high-status, well-compensated knowledge work could prove "politically explosive," as Kinder puts it, even if most jobs survive. The workers facing the greatest near-term risk are precisely those who benefited most from previous technological transitions—a demographic with significant political and economic influence. How policymakers manage this inversion will shape both the AI transition and broader social stability.

Kinder announced during the interview that she is leaving Brookings after three years to launch a new organization focused specifically on solving AI transition challenges. The conversation was first reported by Platformer, with additional details available in their full podcast interview.