Are Women's Jobs Most at Risk From AI? Here's the Data

If you've ever had that flicker of dread about whether your job is safe, here's an honest answer: if you work in an administrative, clerical, or customer-service role, you're standing closer to AI's blast radius than most, and you're more likely to be a woman when you're standing there. On August 19, New York Governor Kathy Hochul made that official, launching the first listening sessions of a new state commission built specifically to study — and try to soften — how AI is reshaping work for women first.

Is AI job loss really worse for women?

The research backs up why Hochul started here. A joint analysis from the Brookings Institution and the Centre for the Governance of AI, using labor data from Lightcast, found that among the roughly 6 million U.S. workers most exposed to AI and least equipped to pivot into something else, 86% are women — concentrated in exactly the clerical and back-office roles now under the microscope in New York. Brookings senior fellow Mark Muro put it plainly: "It's a bit more about what women do in the economy rather than what they are." This isn't about women being less capable with AI. It's about which jobs got handed to women in the first place, and those happen to be the ones AI automates fastest.

That's the exact math behind Hochul's new FutureWorks Commission, a 20-member state panel co-chaired by former U.S. Labor Secretary Tom Perez, TIAA CEO Thasunda Brown Duckett, and Breakwater Initiative founder Molly Kinder. Its first move wasn't a task force full of tech executives. It was a room full of women in the administrative, clerical, and customer-service jobs that make up 84% of New York's back-office workforce. "As New York continues to lead the nation on AI and tech innovation," Hochul said at the launch, "I'm fighting to ensure that progress doesn't come at the expense of working New Yorkers." The commission is due to hand over policy recommendations by the end of 2026.

Why this actually matters for you

If you're in your twenties or thirties working reception, scheduling, billing, claims processing, or customer support, this isn't an abstract policy story — it's your career runway. These jobs have historically been a real, if underpaid, path to stability: predictable hours, a foot in the door, sometimes a path up. Losing that pathway without anything built to replace it doesn't just cost a paycheck, it costs the years of experience you were counting on to leverage into something better. That's exactly the gap you can see playing out in why the AI industry still hasn't fixed its gender gap and in how women keep getting left out of the AI hiring boom — the jobs disappearing and the jobs being created by AI are not landing on the same people.

One thing to try this week

Don't wait for a commission report to find out where you stand. Ask your manager directly which AI tools are being piloted for your role and on what timeline — vague answers are information too. Then spend twenty minutes learning the AI tool closest to your actual job (the scheduling assistant, the customer-service chatbot, the billing software's new "smart" feature) so you're the one who knows how to run it, not the one it's quietly practicing on. And if you've ever held back from mentioning you use AI at work, know you're not imagining that reaction — it's real, and it's worth pushing past, the same way it comes up in why women get judged more harshly for using AI at work.

Quote to sit with

"The most dangerous phrase in the language is, 'We've always done it this way.'" — Grace Hopper, computer scientist and Navy rear admiral whose work laid the foundation for modern programming.

New York is finally treating "we've always done it this way" as the risk it is, at least for one workforce. The rest of us shouldn't wait for a commission to start asking the same question about our own jobs.

💌 Sarah