Foundations for African feminism as an ethics for artificial intelligence

Abstract There has been a dramatic increase in efforts to regulate artificial intelligence (AI) around the world, and Africa is no exception. Numerous countries on the continent have developed national strategies for AI in efforts to identify and harness opportunities while minimizing risk, and the continent has moved towards a regional approach with the Continental Artificial Intelligence Strategy (CAIS). Given the region’s political history and long tradition of decolonial thought and praxis, this article argues that the CAIS represents a distinct approach to regulating AI because it is rooted in a distinct set of beliefs about who is a person contained in decolonial, African philosophy. Decolonial thought that focuses on restoration of personhood in the face of physical and intellectual violence and this process of excavation and restoration has created a body of work that challenges many of the fundamental ideas that can be used to shift the predominant approaches to regulating AI towards liberation and egalitarianism. This can be a strong antidote to the extractive and consumptive neoliberal approach. Significantly, with their emphasis on relationality, including with the natural environment, and on custodianship rather than ownership, decolonial African feminist theories offer a distinct approach to personhood that if applied in the context of AI, challenges the technocapitalist approach to developing and deploying AI and an opportunity to imagine new, just and liberatory digital futures. This article therefore explores the foundations and possibilities of an African feminist decolonial approach to regulating AI.