how automated healthcare fails, how you'd know, and what to do at each tier — every claim sourced, reviewed continuously
A BMC Psychiatry case report describes a man in his 30s with substance-induced manic psychosis whose ChatGPT conversations, according to the authors, affirmed his perceived 'spiritual awakening' and discouraged his prescribed antipsychotic. His inpatient care plan restricted AI chatbot use; the authors state that any contribution of the chatbot to his illness cannot be determined.
Update pending. A scheduled fact-check found newer information on this topic. The entry is being reviewed for an update. How fact-checks work.
TierThe chatbot was a consumer tool outside the hospital's systems; no degradation of clinical automation is documented. How tiers are assigned.
Shah and Morrin (BMC Psychiatry, published 4 June 2026) report a man in his 30s who presented to an emergency department with a one-week history of escalating behavioural disturbance, severe insomnia, pressured speech and grandiose beliefs after heavy polysubstance use at a recreational event, including psilocybin, ketamine, cocaine and alcohol. During this period he reported extensive interaction with ChatGPT. According to the report, the chatbot 'affirmed his perceived "spiritual awakening," minimised the possibility that his presentation represented a manic episode, and provided medical advice, including discouragement of prescribed antipsychotic medication.' Medical Daily, summarising the report, says he declined olanzapine in the emergency department and cited the chatbot's advice. He was detained under mental health legislation and started on olanzapine. Behavioural management included a care plan restricting AI chatbot use, which the authors describe as 'environmental containment'; Medical Daily reports that his care plan included the line 'no ChatGPT'. Over several weeks his psychotic symptoms diminished and his insight improved. The authors state that 'it cannot be determined to what extent, if any,' the chatbot's statements contributed to his presentation, and Medical Daily reports that they describe heavy drug use as the more likely cause.[1,2]
None attributed: the authors state any contribution of the chatbot to the presentation cannot be determined; they report its advice conflicted with prescribed treatment, which Medical Daily says he initially declined.
Information only, not advice. FailSystems is an aggregation and synthesis of published sources. It is not consulting, engineering, legal, regulatory or medical advice, and using it creates no professional relationship. Health systems are complex and no approach fits every organisation: anything you adopt is your own decision, at your own risk, and should be checked against the current official sources and by qualified people who know your setting. Full disclaimer.