The Infra-Souveraen project aims to sharpen the perspective of AI as a distinct form of infrastructure. AI is therefore examined both conceptually and empirically as infrastructure from the perspectives of social sciences, legal studies, ethics, and information theory, and the findings are made applicable for practice. Particular attention is given to the design possibilities of AI systems with regard to digital sovereignty and informational self determination, as structural dependencies on dominant market actors already exist today. The results will be translated into design requirements for AI infrastructures as well as concrete recommendations for business, policy makers, regulatory authorities, and civil society. In addition, proposals will be developed to harmonize approaches to data related risk assessment.
By focusing on the infrastructural effects of AI systems, it becomes possible to identify design options that go beyond mere technical design due to their systemic character. In this way, the project strengthens the foundation for a sovereign and self determined use and design of AI infrastructures and enhances the digital resilience of society.
