Reese Witherspoon recently took to Instagram to teach her millions of followers that women are not using artificial intelligence enough. He ends the video with a call to action to learn AI with him.
Witherspoon received thousands of comments criticizing her for being tone-deaf on issues including environmental impact, data center slowdowns and inherent biases in AI.
Witherspoon is right – there is a growing gender gap with AI – but the answer he got may be more indicative why. If Legally Blonde’s Elle Woods can pass for the millennial woman, who can?
There is no female Altman
Research tells us that women are more dangerous than men. Women always see AI as dangerous, especially if its economic and social consequences are uncertain.
Women are also underrepresented in technology. As of 2025, women will make up nearly a quarter of the global technology workforce, and less than a fifth in senior leadership roles. Men only apply for jobs when they reach 60% of their qualifications, while women tend to apply only when they reach 100% of their qualifications, according to the study.
The need for gender diversity in technology is one part of the discussion. How the use of AI plays out across the public sector is another. A Harvard Business School meta-analysis from April, of 18 different studies covering more than 143,000 people and 25 countries, found that women are 22% less likely to use productive AI than men.
In addition, Deloitte’s November 2024 study found that AI adoption declines with age, and the gender gap is more pronounced in the 45-and-over age group.
Recruitment firm Randstad released a report last year that found 71% of workers with AI skills are male and 29% are female – a gender gap of 42 points. Men are more likely to be offered AI training by employers (35% vs. 27%), and women are underrepresented in AI manufacturing skills (69% vs. 31%).
Michael Morris, global head of platform and talent at Randstad, tells CNET that women are a huge reservoir of underutilized talent in the professional workforce.
“Skills embedded in risk roles, including relationship management, operational coordination, behavioral reasoning and stakeholder communication, are exactly what the new AI roles require,” Morris said. “The connection between those two facts should be obvious.”
But developing skills is not easy for some women, especially mothers. Morris says employers need to make learning AI part of the job, not an after-hours expectation, when many women tend to have more responsibilities than their male counterparts.
“Organizations making the most progress are treating AI literacy as a workplace responsibility instead of an individual,” he says.
While women remain under-represented in core technology roles, jobs around AI can be inclusive options. These include AI ethics and governance, AI in healthcare, education and customer experience, product and operational roles including AI implementation and data analysis focused on social impact.
‘space of time’
According to Lakma Algewatthage, a lecturer in business administration at the Australian Institute of Business, the gender gap in AI adoption is a combination of access, socialization, stereotypes and organizational support.
“The time gap is an important and often underestimated contributor to the topic. Women continue to hold a disproportionate share of care and household responsibilities, leaving little flexible time to experiment with emerging technologies, attend training or build confidence through trial and error,” said Algewatthage.
“AI literacy and discovery is not developed through one-time exposure. It requires continuous practice, curiosity and reflection, all of which take time.”
In her role as a teacher, Algewatthage says she looks for women who are very focused on using AI to produce results and are very interested in understanding how to use it effectively and authentically in their professional contexts.
“This shows a strong desire not only to embrace AI, but to develop the judgment and confidence to use it in a meaningful way,” Algewatthage said.
When women learn to carefully evaluate and use AI responsibly, rather than simply learning to use it, they become more confident and willing to engage with technology.
“Ultimately, closing the gender gap is about redesigning the learning experience to encourage critical judgment, ethical decision-making and lifelong adaptability,” Algewatthage said.
It takes a village
Community-based learning, flexible access and tangible role models are key factors in supporting more women in AI – as we have seen in other fields such as STEM and finance.
Elizabeth Ngonzi, a senior AI consultant, board member of the American Society for AI and assistant professor at New York University, says there is a level of urgency to close the gap before it becomes a major economic divide.
“Employers should build AI learning experience during paid work, not consider it as an additional burden,” said Ngonzi. “Leaders should make access to tools, training and use cases a part of workforce development, especially in jobs where women are concentrated.”
This may look like designing a discovery close to a real workflow, so testing sounds 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 that they can do at work,” he said.
Kandis Tagliabue, a technology professional and mother of five, says women don’t need to do things the way men do. “It is a great privilege that men do not realize they have, but women should not feel discouraged.
Tagliabue is a newly created family OS designed to reduce the mental burden of family responsibilities.
Leticia Mooney, a working mom who also homeschools her son, uses AI to help generate activity ideas and learn cheat sheets. “That was helpful, as I have a limited view of the world and I don’t have time to brainstorm,” he said.
But Mooney says beyond menu planning, shopping lists and schedules, AI is a disruption to workflow.
See the full history
