A California man alleges that ChatGPT's medical advice nearly killed him. Scott Winters is suing OpenAI, claiming the chatbot misdiagnosed him with dysautonomia, and he later suffered a massive pulmonary embolism that required emergency surgery.
The lawsuit, filed in a California state court, marks the latest legal challenge against artificial intelligence companies over the reliability and safety of their products. Winters claims he used ChatGPT to research his worsening symptoms—shortness of breath, chest pain, and fatigue—and the chatbot confidently told him he had dysautonomia, a disorder of the autonomic nervous system. Relying on that diagnosis, he delayed seeking immediate emergency care.
Weeks later, Winters collapsed at home and was rushed to the hospital, where doctors discovered a massive pulmonary embolism—a life-threatening blood clot in his lungs. He underwent emergency surgery and survived, but the lawsuit alleges that ChatGPT's erroneous advice directly contributed to the delay in treatment and increased the risk of death.
The case raises urgent questions about the responsibility of AI companies when their tools dispense medical information. ChatGPT, like many large language models, is not designed to provide medical diagnoses. OpenAI's terms of service explicitly state that the platform should not be used for medical advice. Yet millions of users turn to AI chatbots for health answers, often treating them as authoritative sources.
This is not the first time ChatGPT has been criticized for medical inaccuracies. Researchers have documented that AI models can hallucinate symptoms, cite fake studies, and miss critical red flags. The American Medical Association has called for stricter oversight of AI in healthcare, noting that chatbots lack the contextual understanding and accountability of human doctors.
Winters seeks damages for medical expenses, pain and suffering, and punitive damages, arguing that OpenAI failed to implement sufficient safeguards to prevent harm. The company has not yet filed a response, but in prior lawsuits it has argued that users are aware ChatGPT is an experimental tool and should not rely on it for critical decisions.
If the case proceeds, it could set a precedent for how courts view AI-generated advice in sensitive fields like medicine. Regulatory bodies including the FDA are already exploring whether AI health apps should require clearance. This lawsuit may accelerate those efforts, forcing companies to add stronger warnings or real-time medical disclaimers.
For now, Winters’ ordeal serves as a stark warning: AI chatbots can be helpful, but they are not doctors. The outcome of this legal battle could determine just how much responsibility—and liability—tech giants bear for the words their algorithms produce.
AI
↓ Negative
Man Sues OpenAI, Alleging ChatGPT Gave Him ‘Dangerous Medical Advice’
Scott Winters alleges ChatGPT misdiagnosed him with dysautonomia. He later suffered a “massive pulmonary embolism,” according to the lawsuit.
Corin Cesaric-Epple
CNET
3 min read
6/10
Key Takeaways
- Scott Winters filed a lawsuit in California state court alleging ChatGPT misdiagnosed him with dysautonomia after he described symptoms including chest pain and shortness of breath.
- Winters later suffered a massive pulmonary embolism that required emergency surgery, which he claims was worsened by relying on the chatbot's incorrect diagnosis.
- The lawsuit seeks compensatory and punitive damages from OpenAI, though the exact amount has not been disclosed.
- OpenAI's terms of service explicitly disclaim medical advice, but the suit argues that the company did not do enough to warn users or prevent harmful outputs.
- This case is part of a growing wave of litigation holding AI developers accountable for real-world harm, following similar lawsuits over defamation, copyright, and privacy violations.
Frequently Asked Questions
The lawsuit claims that ChatGPT misdiagnosed Scott Winters with dysautonomia after he described his symptoms. Relying on that diagnosis, he delayed emergency care and later suffered a massive pulmonary embolism, which he says caused severe harm.
ChatGPT is not designed or licensed to provide medical advice. OpenAI's terms of service explicitly state the tool should not be used for medical diagnoses. However, users often treat AI responses as authoritative, leading to potential harm.
Dysautonomia is a disorder of the autonomic nervous system that can cause symptoms like fainting, heart rate abnormalities, and fatigue. It is a legitimate condition but requires proper medical evaluation—not a chatbot diagnosis.
Studies have shown that large language models like ChatGPT can produce inaccurate or fabricated medical information, including citing nonexistent studies. Errors vary by condition, but the risk underscores the need for oversight.
If the court finds OpenAI liable, it could push regulators like the FDA to classify health-related AI as medical devices, requiring premarket approval. It may also force AI companies to implement stronger warnings and real-time disclaimers.
Yes, OpenAI has been sued for defamation (when ChatGPT falsely accused a man of embezzlement), copyright infringement, and privacy violations. This is the first prominent lawsuit over medical misdiagnosis.
Topics
Original source
www.cnet.com
Discussion
Join the discussion
Sign in to post a comment or reply.
No comments yet. Be the first to share your thoughts!