What would AI designed with feminist commitments look like? This is not a utopian question. Feminist scholars and practitioners have developed concrete design principles, governance frameworks, and organisational practices that constitute a genuine alternative vision.

Care ethics as a framework. Care ethics, developed by Carol Gilligan and Nel Noddings, proposes that moral reasoning should centre relationships, context, and responsibility for others, not abstract principles applied universally. Applied to AI governance, care ethics asks: what responsibilities do AI developers have to the specific people their systems affect? How do those responsibilities change when systems affect vulnerable people, the sick, the poor, the incarcerated, children? A care ethics framework for AI would embed ongoing responsibility for impact, not just one-time compliance.

Feminist design principles. These include: designing for the margins rather than the centre (if a system works for the most vulnerable users, it works for everyone); including affected communities in design with genuine decision-making power, not just consultation; making the assumptions embedded in design explicit rather than invisible; building in contestability so that people can challenge algorithmic decisions that affect them.

Labour justice in AI production. Feminist AI governance attends to the supply chain of AI, the data labellers, content moderators, and care workers whose labour makes AI possible. It demands fair wages, safe conditions, mental health support, and labour rights for all workers in the AI production chain, not just the engineers.

Intersectional impact assessment. Impact assessments that evaluate AI systems not just for average performance across demographic groups but for performance at intersections, how does the system perform for Black women? For elderly women of colour with disabilities? For single mothers in poverty? Intersectional analysis reveals harm that aggregate metrics conceal.

Feminist AI governance structures. Including women, and especially women from marginalised communities, in AI governance not as tokens but as decision-makers. Creating governance structures that are accountable to affected communities, not just to shareholders and regulators.

These commitments are not merely aspirational. They are analytically grounded in feminist theory and empirically supported by evidence that diverse, inclusive design processes produce better, fairer, more robust AI systems.

Feminist theory’s most fundamental contribution to AI ethics is this: there is no neutral standpoint. Every AI system reflects the values, assumptions, and interests of the people who built it. Making those values, assumptions, and interests visible, and subjecting them to critical scrutiny, is not a political act. It is an epistemic requirement for any AI that claims to be responsible.