Suche
As digital platforms have become more integral to not just how we live, but also to how we do politics, the rules governing online expression, behavior, and interaction created by large multinational technology firms—popularly termed ‘content moderation,’ ‘platform governance,’ or ‘trust and safety’—have increasingly become salient sites of international political contestation. This project seeks to conceptually and empirically advance our understanding of both public and private rulemaking in and around social networks and other important online platforms, and push the boundaries of work at the intersection of technology and global governance, bridging political science, law, and media studies.
Every time someone trains, deploys, or queries an ‘artificial intelligence’ system, whether they know it or not, they are implicating the practices and services of an enormous number of interlinked businesses in the AI ‘stack’. These actors — data generators, collectors and labellers; model developers, finetuners and hosting marketplaces; and cloud hardware, compute and storage providers —have a huge impact on the prevention (or not) of emerging AI-related harms, such as the generation of synthetic child sexual abuse material, the leakage of personal data from foundation models, and the creation of sophisticated fake content by scammers and confidence tricksters.
Das PROSOCIAL-Projekt ist ein großes Kooperationsprojekt, das von Leibniz von 2025 bis 2028 gefördert wird. Zu den beteiligten Institutionen gehören unter anderem das RWI Essen, die Universität Hamburg, das PIK und viele einzelne Expertenforscher wie Mark Andor (RWI), Andreas Lange (U Hamburg), Lorenz Goette (NUS), René Bekkers (VU Amsterdam), Steffen Huck (WZB), Isabel Thielmann (MPI), Linus Mattauch (PIK & TU Berlin) und andere.