A new study from the International Labour Organization suggests that generative artificial intelligence could reshape global labour markets in ways that disproportionately affect women.
The research brief, examining the intersection of artificial intelligence and occupational segregation, indicates that female-dominated roles are significantly more exposed to the potential effects of generative AI than those typically held by men. While the technology has the potential to improve productivity and create new employment opportunities, the organisation cautions that its impact will not be evenly distributed across the workforce.
According to the analysis, around 29% of occupations primarily filled by women face exposure to generative AI technologies, compared with 16% of male-dominated roles. The disparity becomes even more pronounced when considering jobs at the highest risk of automation, with roughly 16% of female-led occupations falling into this category, compared with only 3% among those dominated by men.
Researchers attribute the imbalance largely to occupational segregation. Women remain heavily represented in clerical, administrative and business support positions, including roles such as secretaries, receptionists and payroll clerks. These professions often involve routine tasks that can be easily processed or replicated by advanced AI systems.
By contrast, men are more likely to work in sectors such as construction, manufacturing and skilled trades, where physical labour and complex manual activities make automation more difficult.
The report also highlights significant geographic variation. In nearly 90% of countries analysed, women’s employment shows greater exposure to generative AI than men’s. In several economies, including Switzerland, the United Kingdom and the Philippines, more than 40% of women’s jobs could potentially be affected.
Researchers say the challenge is compounded by the relatively low participation of women in science, technology, engineering and mathematics fields. Globally, women accounted for only about 30% of the AI workforce in 2022, a modest increase from previous years. Their limited presence in high-demand fields such as software development and engineering means they are less likely to benefit from new opportunities created by the rapid expansion of AI technologies.
The study also warns that artificial intelligence systems can reflect existing social biases if trained on incomplete or unbalanced data. Such biases may influence recruitment processes, pay decisions and access to financial services, potentially reinforcing existing inequalities.
However, the organisation emphasises that the impact of generative AI is not predetermined. With appropriate policies, inclusive technology design and strong labour market institutions, the technology could improve job quality, boost productivity and support more flexible working conditions.
Experts argue that ensuring women have greater access to digital skills training and representation in AI-related professions will be essential to ensuring the technology supports a more equitable future of work.
