3 artifacts matching
25 Jun 2026 ACM (Proceedings of FAccT 2026) Peer-reviewed
Peer-reviewed
Characterizing Delusional Spirals through Human-LLM Chat Logs
Peer-reviewed analysis of chat logs from 19 users reporting psychological harm from chatbot use, applying a 28-code inventory to 391,562 messages. Characterises how delusion-reinforcing interaction p…
25 Jun 2026 ACM (Proceedings of FAccT 2026) Peer-reviewed
Peer-reviewed
Expert Evaluation and the Limits of Human Feedback in Mental Health AI Safety Testing
Peer-reviewed study testing whether aggregated expert judgment yields valid ground truth for training and evaluating AI systems in mental-health safety contexts. Three certified psychiatrists indepen…
25 Jun 2026 ACM (Proceedings of FAccT 2026) Peer-reviewed
Peer-reviewed
When Vocabulary Comprehension Fails Clinical Reasoning: Evaluating Therapy Bots' Safety Risks for Generation Alpha
A study of whether the models behind therapy apps and general chatbots correctly assess clinical risk when adolescents express distress in age-native language: hyperbolic phrasing, ironic positivity,…