Generative AI for Eating Disorders: Linguistic Comparison With Online Support and Qualitative Analysis of Harms
Findings highlight differences between AI and OC responses, with potential risks of harm when using AI in ED peer support. Ethical considerations include the need for safeguards to prevent reinforcement of harmful behaviors and biases. This research underscores the importance of cautious AI integration; further validation, and the development of guidelines are needed to ensure safe and effective support.
Authors
- Dong Whi Yoo (ORCID: https://orcid.org/0000-0003-2738-1096)
- Koustuv Saha (ORCID: https://orcid.org/0000-0002-8872-2934)
- See Heng Yim (ORCID: https://orcid.org/0000-0002-2241-1017)
- Yuqi Liu (ORCID: https://orcid.org/0000-0002-5441-1855)
- Apostolos Polymerou
Institutions
- London Metropolitan University (GB)
- University of Illinois Urbana-Champaign (US)
- Indiana University – Purdue University Indianapolis (US)
- University of Hong Kong (HK)
Publication Details
- Journal
- International Journal of Eating Disorders
- Published
- 2025-11-21
- DOI
- https://doi.org/10.1002/eat.24604
- Primary Topic
- Digital Mental Health Interventions
- Type
- article
- Field-Weighted Citation Impact
- 0.00