Artificial intelligence is increasingly heralded as a key part of humanity’s future. But we need to recognise that, as things currently stand, AI risks leaving women behind. Right now, men far outstrip women in terms of AI industry talent (especially at the senior management level), AI research and even consumer AI adoption.
This is concerning in light of the hopes that so many women have for AI to serve as a levelling – if not empowering – force for them in the marketplace. Will AI accelerate women’s rise to leadership, or will it reinforce the very barriers we seek to dismantle? The answer, as with most transformative technologies, is that it depends entirely on the choices we make today.
The opportunities AI can present to level the playing field for women in leadership are clear. Consider the recruitment process – traditionally a gatekeeping mechanism where unconscious bias has long limited women’s advancement. AI-driven screening tools, when properly designed, can evaluate candidates based purely on qualifications and potential, bypassing the subtle prejudices that have kept talented women from corner offices and boardrooms.
The numbers are encouraging. Studies show that well-designed AI recruitment systems have successfully increased the hiring of female managers and reduced gender discrimination in leadership selection. By focusing on competencies rather than stereotypes, these tools can identify leadership potential that human recruiters might overlook.
Beyond hiring, AI offers women flexible ways to skill development through personalised learning platforms, virtual mentorship programmes and global networking opportunities. Yet, perhaps most powerfully, AI platforms can amplify women’s voices in ways previously unimaginable. Data-driven insights can illuminate once-invisible workplace inequities, providing the evidence needed to drive policy change. Advocacy becomes more effective when supported by irrefutable patterns and trends.
However, we must be clear-eyed about the risks. AI is not neutral; it reflects the biases embedded in its training data and the assumptions of its creators. When AI systems learn from historical data that reflects decades of gender inequality, they can perpetuate – and even amplify – those same biases. An algorithm trained on past promotion decisions may learn to replicate discriminatory patterns, presenting them as objective truth.
As the nature of work is transformed due to new technologies, gender-based stereotypes can also pose a risk to women’s advancement. The narrative that women don't belong in technology is both pervasive and demonstrably false. Yet it persists, shaping everything from childhood education to corporate culture. Young girls receive subtle and not-so-subtle messages that technology is a masculine domain, while women in the field face the exhausting reality of constantly proving their technical credibility.
The irony is particularly bitter when we consider history. Ada Lovelace, widely regarded as the world’s first computer programmer, envisioned the potential of computing machines beyond mere calculation in the 1840s – a full century before the first modern computers were built. During the Second World War, teams of women mathematicians laid the groundwork for modern computing. Grace Hopper revolutionised programming with her development of the first compiler. Ida Rhodes, in the 1960s, laid the groundwork for natural language processing.
Yet these contributions have been systematically minimised or erased from popular narratives about technology’s origins. This historical amnesia compounds present-day challenges. When women’s foundational contributions to computing remain invisible, the stereotype that technology is inherently masculine becomes self-reinforcing. Women who might otherwise pursue careers in AI lack role models and the sense of belonging that comes from seeing themselves reflected in the field’s history.
Today’s statistics paint a sobering picture of women’s representation in AI. According to research by Interface, analysing nearly 1.6 million AI professionals worldwide, women comprise only 22 per cent of AI talent globally. The numbers become even more concerning at senior levels, where women occupy less than 14 per cent of senior executive roles in AI.
In academic and research settings, the disparities are equally stark. Women make up only 18 per cent of authors at leading AI conferences, and just 16 per cent of tenure-track faculty who research AI. A Unesco report confirms these findings, noting that women represent only about one third of researchers in science broadly, with AI showing even more pronounced gaps. Interestingly, it has also been suggested that about 23 per cent of those heading up AI innovation are women.
The World Economic Forum reports that the percentage of male graduates in information and communication technologies is 400 per cent higher than women graduates. This pipeline problem perpetuates the cycle: with women making up only 20 per cent of new faculty hires and 20 per cent of AI-related PhD recipients, the numbers aren’t likely to improve soon without intervention.
Perhaps most concerning is the usage gap. Between November 2022 and May 2024, women made up only 42 per cent of ChatGPT’s 200 million average monthly users. The gender gap widens further with smartphone applications, where only 27 per cent of ChatGPT app downloads came from women. Research from Harvard Business School found that across 18 studies involving more than 140,000 people, women’s adoption of AI tools was 10 to 40 per cent smaller than men’s, with the best estimate showing a 25 per cent gap.
This underrepresentation in AI development itself compounds the bias problem. When the teams designing AI systems are predominantly male, blind spots become inevitable. Perspectives that might catch gender bias go unheard. Solutions that could benefit women leaders remain unconsidered. This is not merely about fairness – it’s about the quality and inclusivity of the AI systems being built to shape our collective future.
The automation threat looms particularly large for women. Research indicates that women are overrepresented in roles most susceptible to AI-driven displacement. This vulnerability breeds anxiety, which in turn can create a self-fulfilling prophecy: women who are anxious about AI may be less likely to engage with AI-driven leadership opportunities, further widening the gap. Studies show that women are also more likely than men to question the ethics of using AI tools, which should be seen as a very valuable perspective for the design of AI tools.
The question is not whether AI will shape women’s leadership – it will. The question is whether we will shape AI to support gender equity or allow it to calcify existing inequalities.
This requires deliberate action on multiple fronts. We need to ensure women participate not just as AI users but as AI creators and decision-makers. Educational programmes must actively recruit and support women in technical fields, and mentorship initiatives must connect aspiring female leaders with those who have navigated similar journeys. Women who are already shaping the AI landscape must be platformed, because they deserve it, but also because to increase the number of women in AI we must feel like we might belong.
