Digital agriculture and gender bias. Artificial intelligence, structural inequalities and legal responses

Abstract

The rapid digital transformation of agriculture is reshaping food production systems, rural economies and agri-food governance through the increasing use of artificial intelligence, big data, digital platforms and predictive technologies. Whilst these innovations are often presented as tools capable of enhancing productivity, sustainability and climate resilience, growing scholarship highlights that digital agriculture is not socially neutral. Rather, AI-driven systems risk reproducing and amplifying pre-existing structural inequalities, particularly gender disparities affecting women in rural and agricultural contexts. Despite women’s central role within global agricultural labour systems, persistent inequalities in access to land, credit, technological infrastructures, training and decision-making processes continue to limit their participation in digital innovation ecosystems. This article examines how artificial intelligence systems applied within agriculture may generate discriminatory effects through biased datasets, opaque decision-making processes and automated forms of resource allocation. Because AI models are frequently trained on historical data reflecting male-centred patterns of land ownership, productivity and access to finance, algorithmic systems may indirectly marginalise women farmers and reinforce existing asymmetries within agricultural labour markets and rural economies. The article further explores how digital platforms, automated credit-scoring tools and smart farming technologies may consolidate power imbalances and create new forms of dependency, particularly affecting women-led agricultural businesses and small-scale producers. Against this background, the article analyses the limits of current European regulatory and judicial responses to algorithmic discrimination in agriculture. Whilst the European Union has progressively developed legal frameworks addressing artificial intelligence, data governance and digital accountability, significant gaps remain in recognising the structural and gendered dimensions of algorithmic bias. Existing regulatory approaches often rely on formally neutral standards that struggle to capture the cumulative effects of discrimination embedded within automated systems. The article argues that addressing these challenges requires a broader approach to AI governance capable of integrating gender-sensitive perspectives into technological design, data governance and public policy. Ensuring meaningful inclusion of women within digital agricultural ecosystems is therefore essential not only to prevent the reproduction of historical inequalities, but also to promote socially sustainable and rights-based models of agricultural innovation.