Drawing Lines in Gray Areas: Human and AI Moderation of Online Speech

Social media platforms face a constant challenge: how to deal with comments that are not illegal but still harmful to public debate, such as uncivil remarks or misleading claims. Andrea Wiesner’s work tackled this problem through interviews with professional moderators and large-scale survey experiments with internet users, asking which moderation strategies are seen as fair, effective, and feasible. In this talk, our speaker will also look at the role of AI: while automated tools can help handle large volumes of content, they often struggle with the nuance and context that these gray areas require. Bringing together human judgment and AI systems raises important questions for the future of online discussions and democratic engagement.

About the speaker: 

Dr. Andrea Wiesner is a postdoctoral researcher at the Department of Communication, University of Vienna, and works at the MEDem research infrastructure on democracy data in Europe. Her dissertation examined how online platforms can moderate “gray area” comments—uncivil or misleading posts that remain lawful but may still affect democratic discourse. Her research focuses on internet users’ perceptions of content moderation, online incivility, and the democratic implications of platform governance.