ORIGINAL ARTICLE
Int. J. Public Health
Dual Chain-Mediation of GenAI Chatbots on Loneliness: Perceived Misinformation Exposure and User Trust via Negative Emotions
1. Soochow University, Suzhou, China
2. Fudan University School of Journalism, Shanghai, China
Article metrics
1
Views
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Abstract
Objectives: Mounting concerns over "AI psychosis" highlight the need to examine psychological risks of GenAI chatbots. Loneliness is a critical outcome of prolonged AI use, often mediated by distorted emotional and cognitive processes. This study tests how GenAI chatbots use is associated with negative emotions, which in turn heighten loneliness through two concurrent mechanisms. Methods: We surveyed 516 adults online and applied PLS-SEM with bootstrapped indirect effects and comparative pathway analysis. Results: GenAI chatbots use significantly predicted perceived misinformation exposure (β = 0.318, 95% CI [0.231, 0.409]) and user trust (β = 0.383, 95% CI [0.294, 0.473]). Both pathways increased loneliness via negative emotions, with the information-quality pathway (β = 0.062, 95% CI [0.036, 0.099]) stronger than the emotional-trust pathway (β = 0.023, 95% CI [0.009, 0.040]); overall chain effects did not differ. Conclusions: GenAI chatbots use contributes to loneliness through dual cognitive and emotional mediations. Given emerging AI psychosis risks, interventions should strengthen misinformation recognition and address trust-related vulnerabilities.
Summary
Keywords
GenAI Chatbots, Loneliness, Negative Emotions, Perceived Misinformation Exposure, User Trust
Received
27 August 2025
Accepted
03 March 2026
Copyright
© 2026 Wang and XU. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Chang XU
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.