The Urgency to Close the AI Gender Divide Before It Affects the Pay Gap

Reese Witherspoon recently took to Instagram to lecture her millions of followers about how women aren’t using artificial intelligence enough. She concluded the video with a call to action to learn AI with her. 

Witherspoon received thousands of comments criticizing her for being tone-deaf regarding issues including the environmental impact, data center backlash and biases inherent to AI. 

Witherspoon is right — there is a growing gender divide with AI — but the response she received might be more indicative of why. If Legally Blonde’s Elle Woods can’t get through to the millennial woman, who can? 

No female Altman 

The research tells us that women are more risk-averse than men. Women consistently perceive AI as riskier, especially when its economic and societal effects are uncertain. 

Women are also less represented in tech. As of 2025, women make up roughly a quarter of the global tech workforce, with less than a fifth in senior leadership roles. Men apply for a job when they meet only 60% of the qualifications, whereas women tend to apply only when they meet 100% of them, according to research. 

The need for gender diversity in technology is one side of the conversation. How AI usage is playing out across the societal spectrum is another. A Harvard Business School meta-analysis from April, of 18 separate studies covering over 143,000 individuals and 25 countries, found that women had 22% lower odds of using generative AI than men. 

Furthermore, a Deloitte study from November 2024 found that AI adoption declines with age, and the gender gap is most pronounced in the 45-and-up group. 

Recruitment company Randstad released a report a year ago that found 71% of AI-skilled workers are men and 29% are women — a 42% point gender gap. Men are more likely to be offered AI training by employers (35% vs. 27%), and women are severely underrepresented in generative AI skills (69% vs. 31%). 

Randstad

Michael Morris, the global head of platform and talent at Randstad, tells CNET that women are the largest underutilized talent reservoir in the professional workforce. 

“The skills embedded in at-risk roles, including relationship management, operational coordination, ethical reasoning and stakeholder communication, are exactly what the new AI-era roles require,” Morris says. “The connection between those two facts should be obvious.” 

But upskilling isn’t as simple for some women, especially mothers. Morris says employers need to make AI learning part of the job, not an after-hours expectation, when many women tend to have more responsibilities than their male counterparts. 

“The organizations making the most progress are treating AI literacy as a workplace responsibility rather than an individual one,” he says. 

While women remain underrepresented in core technical roles, AI-adjacent jobs could serve as bridging options. These include AI ethics and governance, AI in healthcare, education and customer experience, product and operations roles involving AI implementation and data analytics with a social impact focus. 

The ‘time gap’

Nitat Termmee/Getty Images

According to Lakma Algewatthage, a lecturer in entrepreneurial management discipline at the Australian Institute of Business, the gender divide in AI adoption is a combination of access, socialization, stereotypes and organizational support. 

“The time gap is a significant and often underestimated contributor to the topic. Women continue to shoulder a disproportionate share of caregiving and household responsibilities, leaving less flexible time to experiment with emerging technologies, attend training or build confidence through trial and error,” Algewatthage says. 

“AI literacy and adoption is not developed through one-off exposure. It requires ongoing practice, curiosity and reflection, all of which demand time.” 

In her role as lecturer, Algewatthage says she observes women are less focused on using AI simply to generate outputs and are more interested in understanding how to use it effectively and authentically in their own professional contexts. 

“This reflects a strong desire not just to adopt AI, but to develop the judgment and confidence to use it meaningfully,” Algewatthage says.

When women learn to critically evaluate and apply AI responsibly, rather than simply learning to operate it, they become noticeably more confident and willing to engage with the technology.

“Ultimately, closing the gender divide is about redesigning learning experiences to foster critical judgment, ethical decision-making and lifelong adaptability,” Algewatthage says. 

It takes a village 

d3sign/Getty Images

Community-based learning, flexible access and visible role models are key drivers to supporting more women in AI – much like we have seen in other fields such as STEM and finance. 

Elizabeth Ngonzi, an executive AI advisor, board member at the American Society for AI and adjunct assistant professor at New York University, says there’s a level of urgency to close the gap before it becomes a larger economic divide. 

“Employers should build AI literacy into paid work time, not treat it as an extra burden,” Ngonzi says. “Leaders should make access to tools, training and use cases part of workforce development, especially in functions where women are concentrated.” 

This could look like designing adoption around real workflows, so experimentation feels useful. 

Ngonzi encourages mothers and caregivers to start using AI for everyday tasks they already manage, such as schedules, pickups and meal planning. 

“That makes AI feel useful right away, helps reduce the mental load and builds confidence they can carry into work,” she says.  

Kandis Tagliabue, a technologist and mother of five, says women don’t need to do things in the way men do. “It’s a huge privilege that men don’t realize they have, but women shouldn’t feel discouraged. They can do it as a hobby.” 

Tagliabue recently built a family OS that’s designed to reduce the mental load of family responsibilities. 

Leticia Mooney, a working mother who is also homeschooling her son, uses AI to help generate activity ideas and learning cheat sheets. “Those have been useful, as I have a limited worldview and no time for brainstorming,” she says. 

But Mooney says beyond menu planning, shopping lists and schedules, AI is an impediment to the workflow. 

Amanda Smith is a freelance journalist and writer. She reports on culture, society, human interest and technology. Her stories hold a mirror to society, reflecting both its malaise and its beauty. Amanda's work has been published in National Geographic, The Guardian, Business Insider, Vice, News Corp, Singapore Airlines, Travel + Leisure, and Food & Wine. Amanda is an Australian living in the cultural center of gravity that is New York City. See full bio