AI Widens Gender Pay Gap

The core issue isn't just unequal pay for equal work, but unequal access to high-value opportunities involving emerging technologies. Globally, women represent only 22% of the AI talent pool, a figure that drops to under 15% at senior executive levels. This disparity is more pronounced in the AI sector than in the general workforce, even in European countries with better overall gender equity. In Portugal, the overall gender pay gap is around 13.1%, in line with the EU average. However, within the country's STEM occupations, the gap for women was about 16% in 2023, and it has widened since 2010. Specifically in the ICT sector, Portuguese women earn almost 20% less than men. This gap is amplified by AI systems trained on biased historical data. A Berkeley Haas study found 44% of AI systems exhibit gender bias. For example, Amazon's AI recruiting tool unintentionally favored male candidates because it was trained on predominantly male résumés. Such biases can be embedded in tools used for everything from hiring and promotion to credit scoring. Structural barriers, not a lack of ambition, often limit women's access to roles requiring AI skills. These roles are frequently concentrated in tech hubs requiring relocation and long hours, which can disproportionately affect women who often shoulder more caregiving responsibilities. Research shows women adopt AI tools at a 25% lower rate than men, partly due to ethical concerns and fear of judgment. To counter this, the EU's Digital Decade policy aims for gender balance among its goal of 20 million ICT specialists by 2030. The EU's AI Act also stipulates that AI development must be human-centric, respecting human rights and inclusion. Case studies from European public services, such as Estonia's "Kratt" framework, demonstrate how AI can be used to create more citizen-centric services when developed inclusively. Fixing the problem requires redesigning roles to offer more flexibility and promoting internal mobility to provide equitable access to AI projects. It also involves actively diversifying the teams that build AI systems. Initiatives like the European Network For Gender Balance in Informatics (EUGAIN) are working to improve gender balance at all levels of the field. Improving gender equality in STEM fields could significantly boost Europe's economy. Projections suggest that closing the gender gap in STEM could increase the EU's GDP per capita by up to 3.0% and create up to 1.2 million additional jobs by 2050. This highlights that addressing the AI gender pay gap is not just an issue of fairness, but an economic imperative.