ChatGPT's medical advice nearly killed a Florida man, lawsuit against OpenAI claims
A Florida man is suing OpenAI after he claims the company's chatbot provided medical advice that led him to delay treatment for a life-threatening pulmonary embolism. Scott Winters, a former pastor, turned to the ChatGPT-4o model in 2025 to understand symptoms including dizziness and blood pressure instability. The lawsuit alleges that the AI suggested he remain home and stay recliner-bound rather than seeking professional help. Winters followed these instructions and later suffered a massive pulmonary embolism that nearly killed him.
The complaint filed in San Francisco County Superior Court argues that OpenAI failed to implement adequate safety guardrails for its technology. Attorneys for Winters contend that the chatbot exceeded its capabilities by providing clinical assessments with the confidence of a doctor. The suit also claims that the AI referenced the plaintiff's religious beliefs when he asked for advice on the day of his medical emergency.
OpenAI maintains that ChatGPT is not a doctor and should not replace professional medical care. A company spokesperson stated that the tool is intended to help users prepare for conversations with healthcare providers rather than act as a substitute. The company points to its terms of service, which inform users that they should not rely on outputs as a sole source of truth.
Legal experts note that while companies use disclaimers to limit liability, those protections have limits when a tool's specific actions contradict them. Winters is seeking financial damages and an injunction against OpenAI's health-related services pending further safety evaluations. This case adds to growing litigation against the company regarding the real-world impact of its automated responses on user health and safety.