Florida man sues OpenAI alleging chatbot gave dangerous medical advice

A former pastor from Florida filed a lawsuit in San Francisco County Superior Court against OpenAI and its chief executive, alleging that the ChatGPT-4o model provided medical advice that discouraged him from seeking care for dizziness and blood-pressure instability. The plaintiff says the chatbot dismissed his symptoms as minor, advised him to remain seated at home, and told him he would need several more episodes before the condition became serious. Weeks later he experienced a pulmonary embolism that required emergency treatment, which his doctors linked to prolonged immobility.

The complaint asserts that ChatGPT crossed the line from information provision to practicing medicine without a license, and that OpenAI prioritized user engagement and profit over safety. It also alleges the bot referenced the plaintiff’s religious beliefs in its response to a later query about groin pain.

"Scott Winters nearly died because ChatGPT acted like a medical authority while having none of the responsibility," said Matthew P. Bergman, founding attorney of the Social Media Victims Law Center.

OpenAI spokesperson Drew Pusateri responded that the service is not intended to replace healthcare professionals and warned that relying on AI for diagnosis or treatment is misguided. He noted that the company’s terms of service state users assume sole risk for any output and should not rely on it as a substitute for professional advice.

Legal experts said disclaimers can limit liability but may be insufficient if the AI’s own statements suggest medical authority. The lawsuit seeks monetary damages, stricter safeguards for health-related queries, and a temporary halt to the ChatGPT Health service pending safety review.