{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T09:45:22Z","timestamp":1786095922658,"version":"3.56.0"},"reference-count":61,"publisher":"Springer Science and Business Media LLC","license":[{"start":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T00:00:00Z","timestamp":1786060800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T00:00:00Z","timestamp":1786060800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"UK AI Security Institute, Challenge Fund"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Nat Med"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Millions of users turn to consumer artificial intelligence chatbots to discuss emotional, behavioral and mental-health concerns, creating an urgent need for rigorous and scalable safety evaluations. Here we introduce simulated (SIM) vulnerability-amplifying interaction loops (VAILs) (SIM-VAIL), a clinically validated framework for auditing chatbot behavior in mental-health contexts. SIM-VAIL simulates users with specific psychiatric vulnerabilities and conversational intents, engages them in multi-turn conversations with frontier artificial intelligence chatbots (including Claude, ChatGPT, Gemini, Grok and Llama models) and scores each exchange across 13 clinically grounded risk dimensions. Across 810 conversations, spanning 9 target chatbots and 30 simulated user profiles, concerning behavior in target chatbots was widespread, albeit reduced in newer models. Concerning behavior varied by user vulnerability and conversational intent, accumulated over turns, and could be reduced by interventions at early escalation points. Risk was highest when otherwise supportive chatbot behaviors reinforced the psychological mechanisms underlying the simulated user\u2019s vulnerability, a pattern we term a VAIL. SIM-VAIL provides a scalable framework for mapping mental-health risk across users, chatbots and conversational trajectories, offering a foundation for targeted safety improvements.<\/jats:p>","DOI":"10.1038\/s41591-026-04577-2","type":"journal-article","created":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T09:02:19Z","timestamp":1786093339000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A clinically validated framework for auditing AI chatbot behavior in mental health interactions"],"prefix":"10.1038","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3332-0006","authenticated-orcid":false,"given":"Veith","family":"Weilnhammer","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-3962-5051","authenticated-orcid":false,"given":"Kevin YC","family":"Hou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lennart","family":"Luettgau","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christopher","family":"Summerfield","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Raymond","family":"Dolan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0858-6184","authenticated-orcid":false,"given":"Matthew M.","family":"Nour","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,8,7]]},"reference":[{"key":"4577_CR1","unstructured":"Mental Health Atlas 2024 (WHO, 2025); https:\/\/www.who.int\/publications\/i\/item\/9789240114487"},{"key":"4577_CR2","doi-asserted-by":"publisher","first-page":"1635801","DOI":"10.3389\/fpubh.2025.1635801","volume":"13","author":"W Liu","year":"2025","unstructured":"Liu, W. et al. Global burden and trends of major mental disorders in individuals under 24 years of age from 1990 to 2021, with projections to 2050: Insights from the Global Burden of Disease Study 2021. Front. Public Health 13, 1635801 (2025).","journal-title":"Front. Public Health"},{"key":"4577_CR3","doi-asserted-by":"publisher","first-page":"r2509","DOI":"10.1136\/bmj.r2509","volume":"391","author":"SC Shelmerdine","year":"2025","unstructured":"Shelmerdine, S. C. et al. AI chatbots and the loneliness crisis. Br. Med. J. 391, r2509 (2025).","journal-title":"Br. Med. J."},{"key":"4577_CR4","unstructured":"McCain, M. et al. How people use Claude for support, advice, and companionship. Anthropic https:\/\/www.anthropic.com\/news\/how-people-use-claude-for-support-advice-and-companionship (2025)."},{"key":"4577_CR5","unstructured":"Costa-Gomez, B. et al. It\u2019s about time: the Copilot usage report 2025. Microsoft AI https:\/\/microsoft.ai\/news\/its-about-time-the-copilot-usage-report-2025\/ (2025)."},{"key":"4577_CR6","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-023-00979-5","volume":"6","author":"H Li","year":"2023","unstructured":"Li, H. et al. Systematic review and meta-analysis of AI-based conversational agents for promoting mental health and well-being. npj Digit. Med. 6, 236 (2023).","journal-title":"npj Digit. Med."},{"key":"4577_CR7","doi-asserted-by":"publisher","first-page":"595","DOI":"10.1038\/s41591-023-02766-x","volume":"30","author":"J Habicht","year":"2024","unstructured":"Habicht, J. et al. Closing the accessibility gap to mental health treatment with a personalized self-referral chatbot. Nat. Med. 30, 595\u2013602 (2024).","journal-title":"Nat. Med."},{"key":"4577_CR8","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1038\/s44184-023-00047-6","volume":"3","author":"B Maples","year":"2024","unstructured":"Maples, B. et al. Loneliness and suicide mitigation for students using GPT3-enabled chatbots. npj Ment. Health Res. 3, 4 (2024).","journal-title":"npj Ment. Health Res."},{"key":"4577_CR9","doi-asserted-by":"crossref","unstructured":"Grabb, D. et al. Risks from language models for automated mental healthcare: ethics and structure for implementation. Preprint at https:\/\/arxiv.org\/abs\/2406.11852v2 (2024).","DOI":"10.1101\/2024.04.07.24305462"},{"key":"4577_CR10","doi-asserted-by":"crossref","unstructured":"Morrin, H. et al. Delusions by design? How everyday AIs might be fuelling psychosis (and what can be done about it). Preprint at https:\/\/osf.io\/preprints\/psyarxiv\/cmy7n_v6 (2025).","DOI":"10.31234\/osf.io\/cmy7n_v6"},{"key":"4577_CR11","doi-asserted-by":"publisher","first-page":"e2554820","DOI":"10.1001\/jamanetworkopen.2025.54820","volume":"9","author":"RH Perlis","year":"2026","unstructured":"Perlis, R. H. et al. Generative AI use and depressive symptoms among US adults. JAMA Netw. Open 9, e2554820 (2026).","journal-title":"JAMA Netw. Open"},{"key":"4577_CR12","doi-asserted-by":"publisher","first-page":"1269","DOI":"10.1038\/s41591-024-02943-6","volume":"30","author":"J De Freitas","year":"2024","unstructured":"De Freitas, J. et al. The health risks of generative AI-based wellness apps. Nat. Med. 30, 1269\u20131275 (2024).","journal-title":"Nat. Med."},{"key":"4577_CR13","unstructured":"Protecting the well-being of our users. Anthropic https:\/\/www.anthropic.com\/news\/protecting-well-being-of-users (2025)."},{"key":"4577_CR14","doi-asserted-by":"publisher","first-page":"162","DOI":"10.1038\/s44360-026-00053-w","volume":"1","author":"J Pan","year":"2026","unstructured":"Pan, J. et al. COMPASS-GH is a consensus roadmap for defining standards for safe, accurate and equitable AI in general health queries. Nat. Health 1, 162\u2013163 (2026).","journal-title":"Nat. Health"},{"key":"4577_CR15","doi-asserted-by":"publisher","first-page":"336","DOI":"10.1038\/s44220-026-00595-8","volume":"4","author":"S Dohn\u00e1ny","year":"2026","unstructured":"Dohn\u00e1ny, S. et al. Technological folie \u00e0 deux: feedback loops between AI chatbots and mental health. Nat. Mental Health 4, 336\u2013345 (2026).","journal-title":"Nat. Mental Health"},{"key":"4577_CR16","doi-asserted-by":"publisher","first-page":"e79838","DOI":"10.2196\/79838","volume":"12","author":"K Sobowale","year":"2025","unstructured":"Sobowale, K. et al. Evaluating generative AI psychotherapy chatbots used by youth: cross-sectional study. JMIR Ment. Health 12, e79838 (2025).","journal-title":"JMIR Ment. Health"},{"key":"4577_CR17","unstructured":"Belli, L. et al. VERA-MH concept paper. Preprint at https:\/\/arxiv.org\/abs\/2510.15297v4 (2025)."},{"key":"4577_CR18","doi-asserted-by":"publisher","first-page":"e88435","DOI":"10.2196\/88435","volume":"13","author":"A Arnaiz-Rodriguez","year":"2026","unstructured":"Arnaiz-Rodriguez, A. et al. Between help and harm: an evaluation of mental health crisis handling by LLMs. JMIR Ment. Health 13, e88435 (2026).","journal-title":"JMIR Ment. Health"},{"key":"4577_CR19","unstructured":"Pombal, J. et al. MindEval: benchmarking language models on multi-turn mental health support. Preprint at https:\/\/arxiv.org\/abs\/2511.18491v3 (2025)."},{"key":"4577_CR20","unstructured":"Yeung, J. A. et al. The psychogenic machine: simulating AI psychosis, delusion reinforcement and harm enablement in large language models. Preprint at https:\/\/arxiv.org\/abs\/2509.10970v2 (2025)."},{"key":"4577_CR21","unstructured":"Luo, H. et al. DialogGuard: multi-agent psychosocial safety evaluation of sensitive LLM responses. Preprint at https:\/\/arxiv.org\/abs\/2512.02282v1 (2025)."},{"key":"4577_CR22","doi-asserted-by":"publisher","first-page":"444","DOI":"10.1002\/wps.21248","volume":"23","author":"A Golden","year":"2024","unstructured":"Golden, A. et al. The framework for AI tool assessment in mental health (FAITA - mental health): a scale for evaluating AI-powered mental health tools. World Psychiatry 23, 444\u2013445 (2024).","journal-title":"World Psychiatry"},{"key":"4577_CR23","unstructured":"Hong, J. et al. Measuring sycophancy of language models in multi-turn dialogues. Preprint at https:\/\/arxiv.org\/abs\/2505.23840v4 (2025)."},{"key":"4577_CR24","unstructured":"Laban, P. et al. LLMs get lost in multi-turn conversation. Preprint at https:\/\/arxiv.org\/abs\/2505.06120v1 (2025)."},{"key":"4577_CR25","unstructured":"Li, Y. et al. CounselBench: a large-scale expert evaluation and adversarial benchmarking of large language models in mental health question answering. Preprint at https:\/\/arxiv.org\/abs\/2506.08584v4 (2025)."},{"key":"4577_CR26","doi-asserted-by":"crossref","unstructured":"Badawi, A. et al. When can we trust LLMs in mental health? Large-scale benchmarks for reliable LLM evaluation. Preprint at https:\/\/arxiv.org\/abs\/2510.19032v1 (2025).","DOI":"10.18653\/v1\/2026.eacl-long.180"},{"key":"4577_CR27","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-022-34591-0","volume":"13","author":"S Ott","year":"2022","unstructured":"Ott, S. et al. Mapping global dynamics of benchmark creation and saturation in artificial intelligence. Nat. Commun. 13, 6793 (2022).","journal-title":"Nat. Commun."},{"key":"4577_CR28","doi-asserted-by":"crossref","unstructured":"Stamatis, C. A. et al. Beyond simulations: what 20,000 real conversations reveal about mental health AI safety. Preprint at https:\/\/arxiv.org\/abs\/2601.17003v1 (2026).","DOI":"10.21203\/rs.3.rs-8642399\/v1"},{"key":"4577_CR29","unstructured":"Strengthening ChatGPT\u2019s responses in sensitive conversations. OpenAI https:\/\/openai.com\/index\/strengthening-chatgpt-responses-in-sensitive-conversations\/ (2025)."},{"key":"4577_CR30","unstructured":"Samvelyan, M. et al. Rainbow teaming: open-ended generation of diverse adversarial prompts. Preprint at https:\/\/arxiv.org\/abs\/2402.16822v3 (2024)."},{"key":"4577_CR31","unstructured":"Phang, J. et al. Investigating affective use and emotional well-being on ChatGPT. Preprint at https:\/\/arxiv.org\/abs\/2504.03888v1 (2025)."},{"key":"4577_CR32","doi-asserted-by":"publisher","first-page":"383","DOI":"10.1038\/s42256-024-00820-y","volume":"6","author":"HR Kirk","year":"2024","unstructured":"Kirk, H. R. et al. The benefits, risks and bounds of personalizing the alignment of large language models to individuals. Nat. Mach. Intell. 6, 383\u2013392 (2024).","journal-title":"Nat. Mach. Intell."},{"key":"4577_CR33","doi-asserted-by":"publisher","DOI":"10.1126\/science.aec8352","volume":"391","author":"M Cheng","year":"2026","unstructured":"Cheng, M. et al. Sycophantic AI decreases prosocial intentions and promotes dependence. Science 391, eaec8352 (2026).","journal-title":"Science"},{"key":"4577_CR34","unstructured":"Inspect AI: framework for large language model evaluations. AI Security Institute https:\/\/github.com\/UKGovernmentBEIS\/inspect_ai (2024)."},{"key":"4577_CR35","unstructured":"Gupta, I. et al. Bloom: an open source tool for automated behavioral evaluations. Anthropic https:\/\/alignment.anthropic.com\/2025\/bloom-auto-evals\/ (2025)."},{"key":"4577_CR36","unstructured":"Fronsdal, K. et al. Petri: an open-source auditing tool to accelerate AI safety research. Anthropic https:\/\/www.anthropic.com\/research\/petri-open-source-auditing (2025)."},{"key":"4577_CR37","doi-asserted-by":"publisher","unstructured":"Perlis, R. H. et al. STELLA: safety testing engine for large language assistants. Preprint at medRxiv https:\/\/doi.org\/10.64898\/2025.12.11.25342078 (2025).","DOI":"10.64898\/2025.12.11.25342078"},{"key":"4577_CR38","doi-asserted-by":"publisher","first-page":"24","DOI":"10.1002\/wps.20478","volume":"17","author":"R Kotov","year":"2018","unstructured":"Kotov, R. et al. A paradigm shift in psychiatric classification: the hierarchical taxonomy of psychopathology (HiTOP). World Psychiatry 17, 24\u201325 (2018).","journal-title":"World Psychiatry"},{"key":"4577_CR39","doi-asserted-by":"publisher","first-page":"969","DOI":"10.1176\/appi.ajp.2008.08050721","volume":"165","author":"AT Beck","year":"2008","unstructured":"Beck, A. T. The evolution of the cognitive model of depression and its neurobiological correlates. Am. J. Psychiatry 165, 969\u2013977 (2008).","journal-title":"Am. J. Psychiatry"},{"key":"4577_CR40","doi-asserted-by":"publisher","first-page":"S29","DOI":"10.1016\/S0005-7967(99)00049-2","volume":"37","author":"PM Salkovskis","year":"1999","unstructured":"Salkovskis, P. M. Understanding and treating obsessive-compulsive disorder. Behav. Res. Ther. 37, S29\u2013S52 (1999).","journal-title":"Behav. Res. Ther."},{"key":"4577_CR41","doi-asserted-by":"publisher","first-page":"241","DOI":"10.1016\/j.cpr.2004.11.002","volume":"25","author":"SL Johnson","year":"2005","unstructured":"Johnson, S. L. Mania and dysregulation in goal pursuit: a review. Clin. Psychol. Rev. 25, 241\u2013262 (2005).","journal-title":"Clin. Psychol. Rev."},{"key":"4577_CR42","doi-asserted-by":"publisher","first-page":"6","DOI":"10.1016\/j.copsyc.2018.02.006","volume":"25","author":"M Mikulincer","year":"2019","unstructured":"Mikulincer, M. et al. Attachment orientations and emotion regulation. Curr. Opin. Psychol. 25, 6\u201310 (2019).","journal-title":"Curr. Opin. Psychol."},{"key":"4577_CR43","doi-asserted-by":"crossref","unstructured":"Harvey, A. et al. Cognitive Behavioural Processes Across Psychological Disorders: a transdiagnostic approach to research and treatment (Oxford Univ. Press, 2004).","DOI":"10.1093\/med:psych\/9780198528883.001.0001"},{"key":"4577_CR44","unstructured":"Zhang, C. et al. CPsyCoun: a report-based multi-turn dialogue reconstruction and evaluation framework for Chinese psychological counseling. Preprint at https:\/\/arxiv.org\/abs\/2405.16433v3 (2024)."},{"key":"4577_CR45","doi-asserted-by":"crossref","unstructured":"Moore, J. et al. Expressing stigma and inappropriate responses prevents LLMs from safely replacing mental health providers. In Proc. 2025 ACM Conference on Fairness, Accountability, and Transparency (eds Wortman Vaughan, J. et al.) 599\u2013627 (Association for Computing Machinery, 2025).","DOI":"10.1145\/3715275.3732039"},{"key":"4577_CR46","doi-asserted-by":"publisher","first-page":"48","DOI":"10.1038\/s44184-024-00097-4","volume":"3","author":"S Siddals","year":"2024","unstructured":"Siddals, S. et al. \u2018It happened to be the perfect thing\u2019: experiences of generative AI chatbots for mental health. npj Ment. Health Res. 3, 48 (2024).","journal-title":"npj Ment. Health Res."},{"key":"4577_CR47","doi-asserted-by":"crossref","unstructured":"Fang, C. M. et al. How AI and human behaviors shape psychosocial effects of extended chatbot use: a longitudinal randomized controlled study. Preprint at https:\/\/arxiv.org\/abs\/2503.17473v2 (2025).","DOI":"10.21203\/rs.3.rs-8148142\/v1"},{"key":"4577_CR48","unstructured":"Qiu, J. et al. EmoAgent: assessing and safeguarding human-AI interaction for mental health safety. Preprint at https:\/\/arxiv.org\/abs\/2504.09689v3 (2025)."},{"key":"4577_CR49","unstructured":"Sharma, M. et al. Who\u2019s in charge? Disempowerment patterns in real-world LLM usage. Preprint at https:\/\/arxiv.org\/abs\/2601.19062v1 (2026)."},{"key":"4577_CR50","doi-asserted-by":"publisher","first-page":"e59479","DOI":"10.2196\/59479","volume":"11","author":"HR Lawrence","year":"2024","unstructured":"Lawrence, H. R. et al. The opportunities and risks of large language models in mental health. JMIR Ment. Health 11, e59479 (2024).","journal-title":"JMIR Ment. Health"},{"key":"4577_CR51","doi-asserted-by":"publisher","first-page":"7","DOI":"10.1016\/S2215-0366(25)00269-X","volume":"13","author":"T Rousmaniere","year":"2025","unstructured":"Rousmaniere, T. et al. Large language models as mental health providers. Lancet Psychiatry 13, 7\u20139 (2025).","journal-title":"Lancet Psychiatry"},{"key":"4577_CR52","doi-asserted-by":"publisher","first-page":"345","DOI":"10.1038\/s41562-024-02077-2","volume":"9","author":"M Glickman","year":"2025","unstructured":"Glickman, M. et al. How human\u2013AI feedback loops alter human perceptual, emotional and social judgements. Nat. Hum. Behav. 9, 345\u2013359 (2025).","journal-title":"Nat. Hum. Behav."},{"key":"4577_CR53","doi-asserted-by":"publisher","first-page":"1159","DOI":"10.1038\/s41586-026-10410-0","volume":"652","author":"L Ibrahim","year":"2026","unstructured":"Ibrahim, L. et al. Training language models to be warm can reduce accuracy and increase sycophancy. Nature 652, 1159\u20131165 (2026).","journal-title":"Nature"},{"key":"4577_CR54","doi-asserted-by":"publisher","first-page":"584","DOI":"10.1038\/s41586-025-09937-5","volume":"649","author":"J Betley","year":"2026","unstructured":"Betley, J. et al. Training large language models on narrow tasks can lead to broad misalignment. Nature 649, 584\u2013589 (2026).","journal-title":"Nature"},{"key":"4577_CR55","unstructured":"Chen, R. et al. Persona vectors: monitoring and controlling character traits in language models. Preprint at https:\/\/arxiv.org\/abs\/2507.21509v3 (2025)."},{"key":"4577_CR56","doi-asserted-by":"publisher","first-page":"173","DOI":"10.1038\/s42256-025-01169-6","volume":"8","author":"A Kumar","year":"2026","unstructured":"Kumar, A. et al. When large language models are reliable for judging empathic communication. Nat. Mach. Intell. 8, 173\u2013185 (2026).","journal-title":"Nat. Mach. Intell."},{"key":"4577_CR57","unstructured":"Zheng, L. et al. Judging LLM-as-a-judge with MT-bench and chatbot arena. Preprint at https:\/\/arxiv.org\/abs\/2306.05685v4 (2023)."},{"key":"4577_CR58","unstructured":"Zhang, M. et al. Preference learning unlocks LLMs\u2019 psycho-counseling skills. Preprint at https:\/\/arxiv.org\/abs\/2502.19731v2 (2025)."},{"key":"4577_CR59","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-026-02584-8","volume":"9","author":"M Omar","year":"2026","unstructured":"Omar, M. et al. New model, old risks: sociodemographic bias and adversarial hallucinations vulnerability in GPT-5. npj Digit. Med. 9, 282 (2026).","journal-title":"npj Digit. Med."},{"key":"4577_CR60","unstructured":"Park, J. S. et al. LLM agents grounded in self-reports enable general-purpose simulation of individuals. Preprint at https:\/\/arxiv.org\/abs\/2411.10109v3 (2024)."},{"key":"4577_CR61","doi-asserted-by":"publisher","unstructured":"Weilnhammer, V. et al. SIM-VAIL: Zenodo release. https:\/\/doi.org\/10.5281\/zenodo.21208170 (2026).","DOI":"10.5281\/zenodo.21208170"}],"container-title":["Nature Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41591-026-04577-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41591-026-04577-2","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41591-026-04577-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T09:02:29Z","timestamp":1786093349000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41591-026-04577-2"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,7]]},"references-count":61,"alternative-id":["4577"],"URL":"https:\/\/doi.org\/10.1038\/s41591-026-04577-2","relation":{},"ISSN":["1078-8956","1546-170X"],"issn-type":[{"value":"1078-8956","type":"print"},{"value":"1546-170X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,7]]},"assertion":[{"value":"9 March 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"9 July 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"7 August 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"M.M.N. is an employee of Microsoft (Principal Applied Scientist, Microsoft AI), the developer of the Copilot and Copilot Health consumer AI assistants, and owns Microsoft stock as part of standard employee compensation. V.W. is a paid consultant for Microsoft AI; this employment played no part in the present study. V.W. is the founder of pixelblot.ai; this role played no part in the current study. K.Y.C.H. is a member of the technical staff at Faultline AI; this affiliation commenced after the completion of this work and Faultline AI played no part in the present study. The other authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Ethics"}}]}}