Amir H. Payberah - 2025-11-08

Yesterday, at our Co-Liberative Computing reading club, we discussed two papers: "The Feminist AI Framework (FAIF)" [1] and "Data Ecofeminism" [2]. Both articles examine AI through feminist perspectives, including intersectional, decolonial, and posthuman approaches. They look beyond the narrow idea of “human-centred” AI and show how this approach often mirrors the same Eurocentric, patriarchal, and capitalist structures that shape our societies. Moreover, they don’t just ask how we can make AI fairer; instead, they propose examining the deeper systems of power and extraction that lie beneath it.

FAIF reimagines AI through a relational and posthuman perspective, seeing technology as part of a shared, living network of relationships. This view is grounded in the concept of situated and plural knowledges, which challenges the illusion of neutrality in technology, highlighting that all knowledge and design emerge from specific contexts, histories, and power relations. Data Ecofeminism complements this by shifting the focus to the material realities of AI; its reliance on energy, water, minerals, and invisible labor, and emphasizing the need to make these hidden costs transparent through accountability, transparency, and environmental awareness.

Both frameworks resist the obsession with scale and speed that dominates much of AI research today. Instead, they imagine smaller, repairable, and community-driven systems that support local autonomy and shared digital commons. They also turn to Indigenous, feminist, and ecological traditions for guidance, offering ways of knowing that value care, mutuality, and balance over domination and extraction.

[1] A Critique of Human-Centred AI: A Plea for a Feminist AI Framework (FAIF), Tanja Kubes, AI & Society, 2025
[2] Data Ecofeminism, Ana Valdivia, ACM FAccT, 2025