Will AI Take Women's Jobs? What the Data Says in 2026

Ellie at Willa, Editorial lead, Willa · Published October 27, 2026 · Updated October 27, 2026

Headlines about AI and jobs tend to speak in averages, and averages hide the part that matters here. The risk is not evenly spread. Three separate research efforts published over the past year, from the International Labour Organization, a team of economists publishing through the National Bureau of Economic Research, and Harvard Business School, converge on the same uncomfortable pattern: the jobs generative AI can most easily absorb are held disproportionately by women, and women are adopting the tools more slowly than men. Both halves of that sentence are fixable. Here is what the data actually says, and what to do with it.

Will AI take women's jobs?

For most women, AI is more likely to change your job than to eliminate it. But the exposure is real, and it is lopsided. An ILO research brief published in March 2026, "Gen AI, occupational segregation and gender equality in the world of work," found that about 29 percent of female-dominated occupations are exposed to generative AI, compared with 16 percent of male-dominated ones. At the sharp end, the split is starker: 16 percent of female-dominated occupations fall into the highest exposure categories, where the technology could take over a large share of the work, versus 3 percent of male-dominated ones.

This is a global pattern, not a quirk of one labor market. Women face more exposure than men in 88 percent of the countries the ILO analyzed. Exposure also climbs with income level: about 41 percent of jobs in high-income countries involve tasks generative AI can perform, against 11 percent in low-income countries. If you work at a desk in a wealthy economy, this research is about you.

The researchers are careful about what exposure means. In most cases it points to transformation, meaning the tasks inside a job get automated while the job itself continues in a new shape. As ILO senior economist and brief co-author Janine Berg put it, "The impact of generative AI on women's jobs is not predetermined." The outcome depends on how employers redesign roles, and on what each of us does in the meantime.

Why are women's jobs more exposed to AI than men's?

The short answer is occupational segregation. Women are concentrated in clerical, administrative, and business support roles: secretaries, receptionists, payroll clerks, accounting assistants. Those jobs are built around exactly the work large language models handle well, such as drafting, summarizing, scheduling, reconciling records, and answering routine questions. Men are more concentrated in construction, manufacturing, and the manual trades, where the work is physical and far harder to automate with software.

Anam Butt, a co-author of the ILO brief, summarized it in one line: "Generative AI is not entering a neutral labour market." The technology did not set out to target women's work. It happens to be good at the kind of work that decades of hiring patterns have steered women into. That distinction matters, because it means the fix is not to wait for the technology to change. The fix is to change your relationship to it.

Which jobs are most at risk, and who can adapt?

A working paper posted to the National Bureau of Economic Research in early 2026, covered by Forbes, asked a more useful question than most studies: of the people in the most AI-exposed jobs, who can actually move? The researchers looked at 37.1 million workers in the jobs most likely to be affected and found that about 26.5 million of them, roughly 70 percent, already have skills that transfer to other roles. That is real cause for optimism.

The remaining 6.1 million face a harder road. They tend to have fewer transferable skills and less savings to ride out a long job search, and they are more likely to live in smaller towns with fewer alternatives nearby. Within the vulnerable clerical and administrative group, 86 percent of workers are women. One caveat: the paper has not yet been peer-reviewed, so hold the exact figures loosely. The direction, though, matches everything else in this article.

What is the AI gender gap, and why does it make this worse?

Here is the bind. The group with the most to lose from AI automation is also using AI the least. A Harvard Business School working paper, "Global Evidence on Gender Gaps and Generative AI," analyzed 18 studies covering more than 140,000 students and workers in countries including the US, Sweden, Mexico, and China. On average, women adopted generative AI at about a 25 percent lower rate than men. The gap held even when access was equal: among roughly 17,000 entrepreneurs surveyed in Kenya, women were about 13 percent less likely to try ChatGPT.

The reasons the researchers found had nothing to do with ability. Women more often worried that using AI was unethical or felt like cheating, and feared being judged as less competent for relying on it. That fear is not irrational; the research notes women are often penalized more harshly when they are seen as lacking expertise. But the practical effect runs backwards. The people facing the most exposure are giving themselves the least practice with the tools most likely to protect them.

If that hesitation sounds familiar, treat closing it as the single highest-value career move available to you right now.

What can you do to stay ahead of AI?

None of this requires a technical background, a certification, or a degree. It requires deciding that the research above is about you, then acting like it. Five moves, in rough order:

1. Use AI on real work, every week. Not a course, and not a demo. Pick one recurring task, such as the status update, the meeting summary, or the first draft, and run it through an AI tool until you know what it does well and where it fails. Fluency comes from repetitions, and the Harvard data says repetitions are exactly where women are falling behind. If you are unsure where to start, our guide to AI skills for non-technical professionals was written for this.

2. Audit your role for exposure. List what you produce in a typical week, mark what AI has made cheap, and find where your judgment still carries the value. Our step-by-step 30-minute AI career audit walks through the whole process, including how to rewrite your resume around what you find.

3. Shift toward judgment work. The tasks that survive automation are the ones that need taste, context, and accountability: deciding what good looks like, handling the exception, owning the client relationship. Volunteer for those tasks now, before your role gets redesigned without you in the room.

4. Make your AI use visible. Quiet competence will not protect you if your manager assumes AI did the work, or worse, assumes you cannot use it. Mention the workflow you automated in your next review. Show a teammate your prompt. Visibility converts private skill into professional reputation.

5. Build a financial cushion. The NBER data is blunt on this point: savings are what separate workers who can hold out for a good next role from workers who have to take whatever comes first. Even a small buffer buys negotiating power.

Key takeaways

• About 29 percent of female-dominated occupations are exposed to generative AI, versus 16 percent of male-dominated ones, according to the ILO.
• Exposure usually means your tasks change rather than your job disappearing, and roughly 70 percent of workers in the most affected jobs have skills that transfer.
• The highest-risk roles are clerical and administrative, where 86 percent of the most vulnerable workers are women.
• Women adopt generative AI at about a 25 percent lower rate than men, which compounds the exposure problem.
• The fastest protection is practice: hands-on use of AI on your real work, every week, made visible to the people who decide your next role.

The numbers in this article describe the market. They do not describe you. Every stat about adoption gaps and exposed occupations is an average, and averages move one person at a time. If you want to close your own gap alongside women doing the same thing, join Willa or come to an upcoming live event. We practice this stuff together, every week.

• women in AI
• AI job loss
• future of work
• AI skills
• career development

Join Willa, the community where women learn AI together.

All posts