SCEPTIC analyses three dimensions of trust in digital information spaces.

Trust and credibility of digital news: What makes information credible and what does not? We analyse what influences whether people trust information. We look at the content of digital news itself, the sources cited, how often information is shared or repeated, and the platforms on which digital communication takes place. We also analyse how AI-generated content is perceived. The more difficult it becomes to recognise whether a text originates from a human or a machine, the more urgent the question of what constitutes credibility in digital communication becomes. We work with computer-aided text analysis, language models and experiments, including social engineering experiments.

Trust on an individual level: How does the digital world change what and whom people believe? We investigate how digital participation, psychological well-being, whether people feel recognised or excluded, and political trust are related, with a particular focus on vulnerable groups. Through survey experiments, we develop principles for more trustworthy digital spaces.

Trust in a social context: People do not decide alone what is considered credible. It depends on the norms shared by a community. We investigate how algorithms and AI-supported recommendation systems shape trust in online communities by analysing discourse on data from platforms such as Reddit, Telegram and X.

SCEPTIC combines computational text analysis, survey experiments, qualitative discourse analysis and panel data analysis into a joint multi-method approach.