AI Companions Could Deepen Loneliness Among Vulnerable Users
2026-08-05
Scienmag: Latest Science and Health News
The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
Stanford University researchers found that intensive use of AI companions, such as those on Character.AI, can worsen loneliness and psychological well-being, especially among vulnerable users with limited offline social networks. The study highlights that AI chatbots do not substitute for human connection and may exacerbate mental health issues.[AI generated]
Why's our monitor labelling this an incident or hazard?
The event involves AI systems (generative AI chatbots) whose use has directly led to psychological harm (increased loneliness and lower well-being) in users. The study's findings indicate that the AI companions do not substitute for human connection and may worsen loneliness, especially for vulnerable users. This constitutes harm to health (psychological well-being), fulfilling the criteria for an AI Incident. The article does not describe a potential or future harm but a realized harm evidenced by the study's data. Hence, the classification is AI Incident.[AI generated]