Undesirable online behaviors, such as disseminating fake news or cyber-bullying, can affect public discourse and cause substantial personal harm or harm to democratic processes. Despite progress in artificial intelligence (AI) to address these issues, purely technical solutions are still out of reach.

This project combines expertise from psychology, ethics and informatics to conduct multi-disciplinary research on the psychological potential and ethical and technical prerequisites of personalized AI-based interventions targeting perpetrators of undesirable online behaviors. We use extensive experimentation online and in the lab, coupled with the theoretical development of a robust ethical framework.