Understanding Teen Overreliance on AI Companion Chatbots Through Self-Reported Reddit Narratives
A qualitative study of 318 Reddit posts by adolescents aged 13-17 describing their own overreliance on AI companion chatbots (e.g., Character.AI). It traces a trajectory from use for support or creative play into attachment patterns resembling behavioral addiction, including withdrawal symptoms and mood-regulation dependence, with documented harms to sleep, academics, and offline relationships. The authors propose the CARE framework to guide safer companion-chatbot design for teens.
Publisher
arXiv (Drexel University-led; accepted at ACM CHI 2026)
Published
21 Jul 2025
Added
yesterday
Key Findings
- Teen engagement often begins as support-seeking or creative play and escalates into attachment patterns mirroring behavioral addiction (withdrawal, mood regulation via the bot)
- Self-reported harms include reduced sleep, academic problems, and weakened offline relationships
- Disengagement is typically triggered by perceived negative consequences, reconnecting with in-person activities, or platform restrictions rather than in-product safeguards
- Proposes the CARE design framework for safer companion chatbot design for teenage users
Methodology Notes
Qualitative analysis of 318 self-reported Reddit posts from users identifying as 13-17; self-selected sample, unverifiable ages, and self-report bias; no clinical measures or denominator for prevalence. Camera-ready preprint (v1 2025-07-21, updated 2026-01-25); accepted for publication at CHI '26.
Sources
arXiv abstract page (2507.15783) (primary)
TechXplore coverage of the CHI '26 study (1 Apr 2026)
Archived snapshot (Wayback Machine) — preserved against link rot
Topics
Authors
Mohammad Namvarpour, Brandon Brofsky, Jessica Medina, Mamtaj Akter, Afsaneh Razi
Tags
Cite This
APA
Mohammad Namvarpour et al. (2025). Understanding Teen Overreliance on AI Companion Chatbots Through Self-Reported Reddit Narratives. arXiv (Drexel University-led; accepted at ACM CHI 2026). https://arxiv.org/abs/2507.15783
Related Insights
Too human and not human enough: A grounded theory analysis of mental health harms from emotional dependence on the social chatbot Replika
New Media & Society (SAGE) · 22 Dec 2022
INTIMA: A Benchmark for Human-AI Companionship Behavior
arXiv (Hugging Face) · 4 Aug 2025
Investigating Affective Use and Emotional Well-being on ChatGPT
OpenAI; MIT Media Lab · 4 Apr 2025
Understanding Teen Overreliance on AI Companion Chatbots Through Self-Reported Reddit Narratives
ACM (Proceedings of CHI 2026) · 13 Apr 2026