Comparing AI chatbot users and non-users in Germany: Reasons for use, addictive AI chatbot use, mental health, and loneliness

Key Findings

• — Sample: 7,312 German adults; 3,729 chatbot users (42.1% men; mean age 42.2, SD 13.3) and 3,583 non-users (42.6% men; mean age 54.3, SD 14.5)
• — Chatbot users reported significantly higher loneliness, depressive, anxiety and stress symptoms and significantly lower life satisfaction than non-users
• — Users engaged with chatbots mainly for informational and practical reasons (information and inspiration, text-related assistance)
• — Frequency of chatbot use to escape negative emotions was positively associated with addictive tendencies
• — Loneliness statistically mediated the association between addictive AI chatbot use and the mental health variables
• — Most observed effect sizes were small; the authors call for experimental and longitudinal work to establish causal direction

Methodology Notes

Cross-sectional online survey of German adults (n=7,312) with self-report measures; users and non-users differ in age (users about 12 years younger on average), so group differences are not adjusted causal effects; effect sizes mostly small. Affiliations: Ruhr University Bochum (Mental Health Research and Treatment Center) and Technische Universität Dresden. Verification route: the ScienceDirect page is walled; title, authors and dates from the Crossref record (DOI created 2026-09-22, print issue dated February 2027), abstract and affiliations from the OpenAlex record. published_date is the Crossref DOI-registration date, which for this journal coincides with online availability; the exact 'available online' date on the publisher page was not readable. OpenAlex lists a publication date of 2026-09-01 (month precision) for the same record; the DOI-registration date (2026-09-22) is used here.

Sources