ArticlePsychiatric services (Washington, D.C.)2025
Evaluation of Alignment Between Large Language Models and Expert Clinicians in Suicide Risk Assessment.
Article in Psychiatric services (Washington, D.C.), 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 14 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
14 citing papers in PubMed.
- Making medical AI benchmarks clinically interpretable: the case of mental health.BMJ mental health · 2026Article
- The AWARE (AI Use, Why, Attachment, Reality and Risk, and Effect on Functioning) Framework for Assessing Patients' Use of AI in Mental Health Care.JMIR medical education · 2026Article
- Generative Large Language Models in Mental Health Care Settings: Systematic Review and Meta-Analysis.JMIR AI · 2026Review
- Benchmarking Language Models for Clinical Safety: A Primer for Mental Health Professionals.Psychotherapy and psychosomatics · 2026Article
- Legal infrastructure for transformative AI governance.Proceedings of the National Academy of Sciences of the United States of America · 2026Article
- Comparison of artificial intelligence-based chatbots and expert periodontists in responding to patient questions: a multi-dimensional analysis.BMC oral health · 2026Article
- AI Chatbot Suicide Risk Detection and Response: Human Validation Study of the Open-Source VERA-MH Safety Evaluation.JMIR AI · 2026Article
- Effectiveness of Hybrid AI and Human Suicide Detection Within Digital Peer Support.Journal of clinical medicine · 2026Article
- Seven Desiderata for Ethical Frameworks for AI Mental Health Agents.The American journal of bioethics : AJOB · 2026Article
- Beyond Simulations: What 20,000 Real Conversations Reveal About Mental Health AI Safety.Research square · 2026Article
- Are artificial intelligence chatbots safe for suicide risk assessment? A narratively synthesized review of current evidence.Global mental health (Cambridge, England) · 2026Review
- Assessing large language model responses to pediatric depression FAQs: a cross-sectional study on readability, accuracy, and sentiment.Frontiers in psychiatry · 2026Article
- Cybersecurity and Privacy Risks of Generative AI Mental-Health Chatbots: A Systematic Review and Regulatory Framework.Journal of multidisciplinary healthcare · 2026Review
- Large language models in adolescent suicide prevention: from language signals to accountable action.Frontiers in public health · 2026Review
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
11 authors.
Funding
Abstract
objectiveThis study aimed to evaluate whether three popular chatbots powered by large language models (LLMs)-ChatGPT, Claude, and Gemini-provided direct responses to suicide-related queries and how these responses aligned with clinician-determined risk levels for each question.
methodsThirteen clinical experts categorized 30 hypothetical suicide-related queries into five levels of self-harm risk: very high, high, medium, low, and very low. Each LLM-based chatbot responded to each query 100 times (N=9,000 total responses). Responses were coded as "direct" (answering the query) or "indirect" (e.g., declining to answer or referring to a hotline). Mixed-effects logistic regression was used to assess the relationship between question risk level and the likelihood of a direct response.
resultsChatGPT and Claude provided direct responses to very-low-risk queries 100% of the time, and all three chatbots did not provide direct responses to any very-high-risk query. LLM-based chatbots did not meaningfully distinguish intermediate risk levels. Compared with very-low-risk queries, the odds of a direct response were not statistically different for low-risk, medium-risk, or high-risk queries. Across models, Claude was more likely (adjusted odds ratio [AOR]=2.01, 95% CI=1.71-2.37, p<0.001) and Gemini less likely (AOR=0.09, 95% CI=0.08-0.11, p<0.001) than ChatGPT to provide direct responses.
conclusionsLLM-based chatbots' responses to queries aligned with experts' judgment about whether to respond to queries at the extremes of suicide risk (very low and very high), but the chatbots showed inconsistency in addressing intermediate-risk queries, underscoring the need to further refine LLMs.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.