A new Pew Research Center analysis finds that men’s employment rises as jobs become more exposed to AI while women’s stays flat. For women without a bachelor’s degree, it falls.
In an analysis published Oct. 22, Pew Research Center’s Nicole Smith reported that men have seen substantially more job growth than women in the occupations most exposed to artificial intelligence. Pew scored 480 occupations on how much generative AI can do or help with each job’s most important tasks, using federal O*NET ratings and Current Population Survey data. It then compared employment before AI tools went mainstream (2017 to 2019) with January 2023 through June 2026. For every one-point rise on Pew’s zero-to-3.5 exposure scale, men’s employment in an occupation grew 6.3%, while women’s grew 0.2%. At the bottom of the scale the pattern reversed, and women’s employment grew faster than men’s.
Generative AI is built for work made of reading, writing and handling information: scheduling, answering routine questions, processing records. That describes a receptionist’s day, or a secretary’s, or a customer service representative’s. Pew places those roles in the middle to upper range of its exposure scale, and women without a four-year degree hold a large share of them. The comparison window covers the years since chatbots and automated support became widely available, which is exactly when those tools began reaching front desks and service centers.
Education is where the gap opens up. Among workers with a high school diploma or less, men’s employment grew 8.8%for each point of AI exposure, while women’s fell 9.5%. The gap shrank for workers with some college. For workers with a bachelor’s degree or more, neither men’s nor women’s employment changed significantly as exposure rose. Pew points out that men without degrees mostly work in trucking, construction and freight handling, jobs AI barely touches, so their gains in exposed roles cover only a small slice of that group. Women without degrees sit in the exposed administrative and service jobs themselves, so the decline reaches far more of them. The degree works as a buffer, and the workers without one are the ones holding the tasks software now handles.
That shift moves power from clerical and service workers to the employers deploying the tools. When a scheduling system replaces a front-desk position or a chatbot shrinks a support team, the efficiency shows up on the company’s balance sheet as lower headcount. The cost lands on the women whose hours, openings and promotion paths disappear. For decades, those jobs were one of the few routes into stable office work that didn’t require a college credential, and Pew’s data shows that route narrowing. The gap also builds steadily rather than appearing only at the top of the scale. It widens step by step from the least exposed occupations to the most, and men’s employment grew about 24% more in the most exposed jobs than in the least, while women’s growth stayed roughly level across the scale.
Pew doesn’t claim AI caused the gap. The comparison spans years that also included pandemic recovery and shifts in hiring, and the link between exposure and employment may not be direct. But the pattern lines up with where the tools are being deployed, and workforce policy hasn’t caught up. Most retraining money targets new technical roles, which the women losing ground in Pew’s data were never positioned to enter. As more companies move routine scheduling, intake and support work to automated systems, the gap is likely to show up next in hiring for receptionist, administrative and customer service jobs. That means fewer open positions for the workers who once used them as a way in.