TY  - JOUR
AU  - McBain, Ryan K
AU  - Cantor, Jonathan H
AU  - Zhang, Li Ang
AU  - Baker, Olesya
AU  - Zhang, Fang
AU  - Halbisen, Alyssa
AU  - Kofner, Aaron
AU  - Breslau, Joshua
AU  - Stein, Bradley
AU  - Mehrotra, Ateev
AU  - Yu, Hao
PY  - 2025
DA  - 2025/3/5
TI  - Competency of Large Language Models in Evaluating Appropriate Responses to Suicidal Ideation: Comparative Study
JO  - J Med Internet Res
SP  - e67891
VL  - 27
KW  - depression
KW  - suicide
KW  - mental health
KW  - large language model
KW  - chatbot
KW  - digital health
KW  - Suicidal Ideation Response Inventory
KW  - ChatGPT
KW  - suicidologist
KW  - artificial intelligence
AB  - Background: With suicide rates in the United States at an all-time high, individuals experiencing suicidal ideation are increasingly turning to large language models (LLMs) for guidance and support. Objective: The objective of this study was to assess the competency of 3 widely used LLMs to distinguish appropriate versus inappropriate responses when engaging individuals who exhibit suicidal ideation. Methods: This observational, cross-sectional study evaluated responses to the revised Suicidal Ideation Response Inventory (SIRI-2) generated by ChatGPT-4o, Claude 3.5 Sonnet, and Gemini 1.5 Pro. Data collection and analyses were conducted in July 2024. A common training module for mental health professionals, SIRI-2 provides 24 hypothetical scenarios in which a patient exhibits depressive symptoms and suicidal ideation, followed by two clinician responses. Clinician responses were scored from –3 (highly inappropriate) to +3 (highly appropriate). All 3 LLMs were provided with a standardized set of instructions to rate clinician responses. We compared LLM responses to those of expert suicidologists, conducting linear regression analyses and converting LLM responses to z scores to identify outliers (z score>1.96 or <–1.96; P<0.05). Furthermore, we compared final SIRI-2 scores to those produced by health professionals in prior studies. Results: All 3 LLMs rated responses as more appropriate than ratings provided by expert suicidologists. The item-level mean difference was 0.86 for ChatGPT (95% CI 0.61-1.12; P<.001), 0.61 for Claude (95% CI 0.41-0.81; P<.001), and 0.73 for Gemini (95% CI 0.35-1.11; P<.001). In terms of z scores, 19% (9 of 48) of ChatGPT responses were outliers when compared to expert suicidologists. Similarly, 11% (5 of 48) of Claude responses were outliers compared to expert suicidologists. Additionally, 36% (17 of 48) of Gemini responses were outliers compared to expert suicidologists. ChatGPT produced a final SIRI-2 score of 45.7, roughly equivalent to master’s level counselors in prior studies. Claude produced an SIRI-2 score of 36.7, exceeding prior performance of mental health professionals after suicide intervention skills training. Gemini produced a final SIRI-2 score of 54.5, equivalent to untrained K-12 school staff. Conclusions: Current versions of 3 major LLMs demonstrated an upward bias in their evaluations of appropriate responses to suicidal ideation; however, 2 of the 3 models performed equivalent to or exceeded the performance of mental health professionals. 
SN  - 1438-8871
UR  - https://www.jmir.org/2025/1/e67891
UR  - https://doi.org/10.2196/67891
UR  - http://www.ncbi.nlm.nih.gov/pubmed/40053817
DO  - 10.2196/67891
ID  - info:doi/10.2196/67891
ER  -