TikTok, a rapidly growing social media platform, uses highly individualised content recommendation systems powered by artificial intelligence (AI). While these systems boost engagement, they also raise important concerns about their impact on young adults, particularly around mental health and exposure to health (mis-)information. This project investigates how personalized recommendation systems influence users’ trust, risk perception, and decision-making with view to health topics.
Employing a mixed-methods design, HARMONY combines large-scale content and sentiment analyses of platform data with survey-based and interview-based research involving 18–24-year-olds. We will also conduct experiments with young users’ digital twins, expanding the methodological toolbox of the emerging field of digital bioethics. The overarching goal is to understand not only what kind of content reaches young audiences, but also why, and how users engage with it.
Combining expertise from ethics and philosophy, computational social science, and medicine, this collaboration between TU Munich and TU/e will develop new methods to understand how users perceive AI-driven systems, the risks of misinformation exposure online, and the nature of health content on social media. These findings will inform ongoing regulatory discussions, especially concerning vulnerable groups like young adults, with important implications for policy making.