The abstract is read at the publisher; the summary is JClub's.
AMHART auditing of four major AI chatbots found guideline-informed response coverage is greater for anxiety than depression, and for moderate than severe adolescent presentations.
- Why it matters: General-purpose chatbots are widely used by adolescents, yet they have limited evaluation against youth-relevant clinical standards.
- What they did: The clinician-informed AMHART instrument was developed across seven domains, with clinician-reviewed items and strong content validity plus inter-rater reliability, then used to score multi-turn conversations.
- The result: A three-model LLM evaluator ensemble agreed with human reference labels and enabled AMHART benchmarking, suggesting a reliable clinician-informed way to evaluate chatbot responses to adolescent anxiety and depression disclosures.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.