By Reshma Kapadia
The rapid development of artificial intelligence has sparked various concerns about its impact on the labor market, society, and business models. Add this one to the list: The potential risk to women is even bigger -- from reinforcing gender biases to hitting roles disproportionately held by women.
A report released this week by JP Morgan's strategic research team highlighted a confluence of factors related to AI that risk widening gender gaps and could dent the promise of AI. The team in a report released this week cautioned that these gaps need to be addressed to "avoid creating a self-reinforcing cycle of biases, underrepresentation and weaker productivity gains."
One potential problem the analysts highlighted: The underrepresentation of women in technology, especially in AI, which reduces their visibility in training data and leadership.
Colleges are graduating more women than men, but they still make up just a quarter of those studying computer science. While women have grown to a third of those working in STEM, a LinkedIn/World Economic Forum report last year found that retention remains a challenge in STEM, with men overrepresented in every career stage, especially the C-suite.
That's especially true in the small cohort of tech workers powering AI's development, which represents about half a percentage point of global workers. Women make up less than a third of AI-skilled workers, according to the report.
Of course, there are notable exceptions, including some on Barron's Most Influential Women in U.S. Finance who are helping shape AI. One notable outlier is Saudi Arabia, where women outnumber men in the AI-skilled talent pool. The country is in the midst of an ambitious effort to transform its economy, diversifying beyond energy into AI and technology more broadly while liberalizing women's rights.
Not only are women often not in the rooms where AI is being developed, they also aren't using AI as much. The economists cited several reports showing a gender gap. in AI adoption and perception. While about half of men reported using generative AI in the prior 12 months, a study by the Stanford Social Innovation Review found that only 37% of women had done so. Another Harvard Business School meta-analysis of 18 separate studies across 25 countries found women had 22% lower odds of using generative AI than men.
Of the roughly 37 million workers in the top quartile of jobs exposed to AI, 26.5 million are in occupations that also have an better-than-median ability to adapt to the technology and relatively well-equipped to handle job transitions if AI gets rid of their roles, according to a report released earlier this year by the Brookings Institution on AI and the workforce.
Jitters about losing jobs to AI are broad-based, but the researchers found that women may be more vulnerable because of the types of roles in which they are overrepresented. For example, of the six million people in the U.S. holding clerical and administrative roles, roughly 86% are women, according to the Brookings report. This group was among the most vulnerable to AI, in part because of the older average age and narrower skill sets of those in the field, as well as fewer alternative opportunities.
Those looking for a new job in the age of AI also face challenges, with JP Morgan analysts warning that AI systems can exacerbate gender biases.
After analyzing 1.4 million images and videos across internet platforms and billions of pieces of text from the web in nine language models, Douglas Guilbeault, a professor at Stanford's Graduate School of Business, and two other researchers found women were consistently described as younger than men in depictions of higher-paid and higher-status roles.
When prompted to create thousands of unique resumes for 54 occupations using typical male or female names, ChatGPT produced results that portrayed women as younger and less experienced, while assigning male applicants higher quality scores, according to the research published in Nature last fall.
These gaps matter not just for the economy but also for investors. If the generative AI adoption gap persists, the JP Morgan economists warn that AI products and services may not reach their full potential; the productivity gains fueling heavy investment in AI could be at risk, and half the population could miss out on the benefits of future technologies, whether in healthcare or elsewhere.
Write to Reshma Kapadia at reshma.kapadia@barrons.com
This content was created by Barron's, which is operated by Dow Jones & Co. Barron's is published independently from Dow Jones Newswires and The Wall Street Journal.
(END) Dow Jones Newswires
March 22, 2026 03:00 ET (07:00 GMT)
Copyright (c) 2026 Dow Jones & Company, Inc.
Comments