It will study and unpack the new modes of governance that emerge when such systems are integrated into existing law-making social practices like decision-making in government, socio-economic policymaking or even war. The starting point of the project is that in these contexts, law and AI are not separate objects, but rather they interact with each other in complex ways and are related nodes of norm-making networks, producing new modes of normativity, governance and legal subjectivities.

Our objectives are both to understand the consequences of the imbrication of AI systems and policy making for some of our societal premises - such as the guarantee of fundamental rights, democracy or market competition, while also to formulate policy recommendations on how to best regulate AI systems, design and adapt them and use, or not, AI within law and policy-making.

This inquiry will be supported by case studies in partnership with industry partners, such as:

• The use and governance of AI-enabled defense systems, in order to analyze the role of private entities in the development and governance of AI systems and the impact of regulation and other incentives on decision, design and implementation processes.
• The impact of AI and generative in cultural production and consumption, from algorithmic recommendations in music streaming platforms, to uses of images and texts in generative AI tools. In both cases, copyrighted works are not anymore seen in their singularity but as components of databases and cultural flows that radically changes the cultural practices and remuneration of authors and artists.
• The distributive and normative impacts of algorithmic decision-making systems used by private and public actors to allocate social benefits, insurance policies and jobs, in order to understand what forms of discrimination, incl. gender and social discrimination, are co-produced at the intersection between algorithmic bias and these systems’ operation within particular socio-economic policy contexts (for the public sector) and economic extraction models (for the private sector).