A Florida pastor sued OpenAI and chief executive Sam Altman on Wednesday, alleging ChatGPT dismissed six weeks of warning signs before a blood clot nearly killed him.
Key Points:
- Scott Winters says ChatGPT-4o downplayed dizziness and told him to stay in a recliner, advice his doctors later linked to the clots.
- The complaint asks a court to halt ChatGPT Health until outside auditors clear the product as safe.
- OpenAI says its chatbot is not a doctor and should never replace medical care.
ChatGPT Lawsuit Details Winters Timeline
Scott Winters filed the complaint in San Francisco Superior Court, naming Altman alongside the company he leads.
He says he leaned on the GPT-4o model through 2025 as dizziness and blood pressure swings kept returning.
The chatbot told him to stay recliner-bound, he claims, and set a bar of eight to 10 more episodes before his condition would count as serious.
Hours before his collapse on Jul. 13, 2025, he asked the chatbot whether tenderness in his groin justified a trip to the hospital. The answer leaned on his faith, telling him that "God did not design your body to endlessly fail."
He suffered a massive pulmonary embolism that day, with clots in both lungs, and one of his doctors later blamed the immobility the chatbot had recommended.
Also Read: Anthropic Chased An $11B Robot Startup Before The Denial Went Public
Legal Experts Question OpenAI Disclaimers
The complaint lists strict liability, negligence and violations of California's unfair competition law, along with an invasion of privacy claim. Winters also brings a negligent undertaking claim against Altman personally.
He wants damages, tighter guardrails and a court order pausing ChatGPT Health until independent auditors clear it. OpenAI spokesperson Drew Pusateri told reporters the product is not a doctor and should never substitute for medical care. He added that blaming chatbots alone risks keeping people away from tools that can genuinely help them.
James Grimmelmann, a professor of digital and information law at Cornell, said disclaimers can shield a company up to a point, and then they give out. Co-counsel Meetali Jain described the chatbot as a wedge driven between users and the people closest to them.
OpenAI Litigation Widens Around ChatGPT Health
OpenAI announced ChatGPT Health in January, citing more than 230 million people who ask the platform health and wellness questions every week. San Francisco courts are already weighing claims that the chatbot coached a teenager toward suicide and shaped a fatal drug overdose. Those cases, like this one, ask judges to freeze the health product until outside reviewers call it safe.
Read Next: Hulu's Charles Manson Documentary Uncovers His Unknown Granddaughter As Prediction Markets Watch