Women are more skeptical than men about artificial intelligence and its adoption in the workplace when the economic outcomes of the technology are uncertain, according to new research from Northeastern University published in the academic journal PNAS Nexus.
Based on survey data from the United States and Canada, the study found that women consistently viewed AI as riskier, particularly when its potential economic effects were unclear.
Beatrice Magistro, an assistant professor of AI governance at Northeastern University and a co-author of the research, said the findings add to existing research on gender differences in attitudes toward technology, with implications for AI policy, adoption and the future of work.
Researchers surveyed about 3,000 people in the U.S. and Canada. They identified two factors that help explain the gender gap: women’s greater tendency toward risk aversion and higher exposure to potential harms linked to AI, including job displacement, bias and widening inequality.
Participants were asked whether they believed the risks of generative AI outweighed its benefits. Researchers also measured respondents’ general willingness to take risks using a lottery-style question. Participants chose between a guaranteed $1,000 payout and a 50% chance of winning $2,000, with a 50% chance of winning nothing. Those who chose the probabilistic option were classified as risk-takers, while those who chose the guaranteed payout were classified as risk-averse.
Respondents rated on a 0-10 scale whether AI’s risks outweighed its benefits. Women were about 11% more likely than men to say AI’s risks outweighed its benefits, Magistro said. She noted that the gap is similar in size to gender differences in attitudes toward trade, which researchers used as a benchmark.
In open-ended responses, women were more likely than men to express uncertainty about AI. Women were 6 to 7 percentage points more likely to say they did not know what AI’s benefits were, and 2 to 3 percentage points more likely to say they were not sure or believed there were no benefits.
Magistro said the data also showed that the gap depends on uncertainty. When AI-driven job gains were guaranteed, men and women responded similarly. When employment outcomes were uncertain, women became significantly more skeptical.
“Basically, when women are certain about the employment effects, the gender gap in support for AI disappears,” Magistro said. “So it really seems to be about aversion to uncertainty.”
Magistro suggested the gap may reflect long-standing patterns of exclusion and socialization, including women’s historical underrepresentation in STEM fields and fewer opportunities for exposure to emerging technologies.
The research noted that women’s underrepresentation in STEM may limit access to high-paying AI-related jobs and leadership roles, potentially widening gender pay gaps. Magistro said policymakers should address gender-specific concerns in AI policy, warning that AI could deepen inequality or trigger political backlash.
