The extraction of ecological value from nature, labor value from workers, and social value from communities constitutes the root cause of pollution, poverty and social domination. Indigenous traditions, commons-based production and related alternatives offer models in which value is not extracted, but rather circulated back to the human and non-human agencies that generated it. In this talk we will describe how this framework of “generative justice” can be used to design decolonized technologies. In Detroit our experiments develop AI and other applications in a platform for community-based economies. In Africa they are developed for a collective of crafters, artists and other creatives. We find that establishing democratized, regenerative value flows requires “full stack decolonization,” because extraction occurs at every layer from infrastructure to machine learning algorithms. The transition to worker-owned systems, in which they determine where AI will be located across the broad spectrum of human-machine agency possibilities, is a key component in developing pathways for egalitarian and sustainable futures.

Ron Eglash is the Director of the Center for Generative Justice and a former Professor in the School of Information at University of Michigan. Audrey Bennett is a Professor in the Stamps School of Art and Design at the University of Michigan and a University Diversity and Social Transformation Professor.