Generative AI (Gen AI) promises to transform economies, workforces, and innovation ecosystems. But this transformation is unfolding along a sharply uneven playing field – especially when it comes to gender. A new World Economic Forum report makes the argument clear: closing the gender gap in AI is not only a matter of fairness. It’s a requirement for technological progress, workforce resilience and economic growth.
Gen AI is reshaping the workforce – but not equally
As Gen AI tools spread across sectors – from software development to legal research to creative production – the distribution of benefits and risks is highly uneven. Women are more likely to be in jobs disrupted by Gen AI, and less likely to be in roles that are augmented by it. According to LinkedIn data from the report, 57% of disrupted roles are held by women, while only 46% of augmented roles are. It’s not just an issue of occupational distribution. Even among professionals with AI skills, women remain underrepresented. In 2025, only 1.1% of women on LinkedIn list AI engineering skills, compared to 2.0% of men. The gender gap is narrowing – but slowly.
Talent pipelines are leaking at every stage
Much of the problem starts early. Women still make up less than one-third of the global STEM workforce. Among female STEM graduates, many never enter tech industries at all. And those who do often face a steady drop in representation as they move toward senior leadership. This phenomenon – the “school-to-work” and “drop-to-the-top” attrition – is a major contributor to the imbalance we now see in AI development. Even in countries with high rates of female STEM graduates, retention and promotion remain weak. In 2024, women held 24.4% of managerial roles in STEM, but only 12.2% of C-suite positions.
Skilling is essential – but insufficient on its own
Gen AI is creating a new kind of digital divide – not just around access to tools, but around readiness for transformation. The report shows that women are less likely than men to expect their roles will change significantly due to AI. They’re also less likely to report confidence in navigating that change. This isn’t just a perception issue. Many skilling initiatives still disproportionately reach male employees. Research from Randstad cited in the report shows that employers are prioritising Gen AI upskilling for men in most countries surveyed.
Yet the potential is there. Between 2018 and 2025, the share of women listing AI skills on LinkedIn has grown substantially. In 74 of 75 economies studied, the gender gap in AI talent has narrowed.
Fair AI starts with fair data and fair systems
One of the report’s more revealing insights concerns AI in recruitment. Today, 99% of Fortune 500 companies use automation in hiring. But if the training data behind these systems reflect past inequities, the systems will reproduce them. There is an opportunity to use AI to make hiring and promotion fairer – identifying overlooked talent, reducing subjectivity in performance reviews, and removing bias from decision-making. But that only happens with intentional design. Without transparency, diverse datasets and rigorous oversight, these systems could entrench the very gaps they claim to close.
AI can help reimagine recruitment, promotion and evaluation processes – but only if we intentionally build systems that detect and counteract bias. This means:
-
designing training datasets that are balanced and representative
-
creating evaluation tools that reward diverse career paths and skill sets
-
using AI not just to sort candidates, but to identify and elevate overlooked talent
The potential here is transformative: AI could become a tool not just for efficiency, but for equity.
Parity is a competitive advantage
Gender-diverse teams perform better. They bring broader perspectives, challenge groupthink, and build more inclusive products. In AI specifically, greater diversity reduces bias in systems, strengthens innovation pipelines, and increases market relevance. Economies with more inclusive talent strategies will be better positioned to scale AI adoption and navigate disruption. Those that neglect gender disparities risk falling behind – not just ethically, but economically.
Recommendations
The report concludes with a clear call to action for leaders across sectors. Whether in business, government, or academia, there are steps to take now to ensure AI delivers broadly shared benefits – not just productivity gains for the few.
1. Expand and diversify AI talent pipelines
– Invest in inclusive STEM education and AI literacy
– Track and close the “school-to-work” and “drop-to-the-top” gaps in women’s careers
– Promote women into AI leadership and strategic tech roles
2. Embed gender parity in Gen AI strategy
– Use data-driven tools to monitor AI’s differential impact on men and women
– Prioritise the inclusion of women in Gen AI application design, especially in high-impact fields
– Ensure female voices shape how AI is deployed in government, healthcare, finance, and beyond
3. Make augmentation equitable
– Design upskilling programmes that specifically target disrupted roles where women are overrepresented
– Provide visibility, mentorship and support to women transitioning into AI-adjacent fields
– Fund AI research that focuses on equitable augmentation and task revaluation
4. Build inclusive AI systems
– Audit AI hiring and evaluation systems for gender bias
– Redesign performance metrics to capture collaborative, communicative and relational work
– Incentivise inclusive outcomes, not just technical accuracy or speed
5. Align industry and policy efforts
– Governments must support gender-responsive AI policy frameworks
– Public-private partnerships should pool data and best practices
– Global forums should track progress with accountability and transparency
A turning point for AI and gender equity
AI is not yet fully entrenched; its systems, institutions and practices are still being formed. That means we still have the chance to shape it. The report’s data shows some promising signs: the gender gap in AI talent is narrowing in 74 of 75 economies studied. But without sustained effort, it could easily reverse. To avoid that, leaders must act with intentionality. Parity in AI isn’t just about DEI or ethical choices, it’s about performance, innovation, and resilience. The future of AI will be shaped by those who build it, so let’s make sure that future is shaped by everyone.