OpenAI Sued as ChatGPT Allegedly Gives Near-Fatal Medical Advice

OpenAI faces a new lawsuit alleging that its AI chatbot, ChatGPT, provided dangerously flawed medical advice, potentially leading to a near-fatal health crisis for a Florida pastor. Scott Winters has filed a lawsuit in California against OpenAI and its CEO, Sam Altman, claiming that ChatGPT-4o offered "extremely dangerous" recommendations that delayed his treatment for a pulmonary embolism.

“

ChatGPT gave him "extremely dangerous" recommendations that delayed the treatment of a potentially fatal pulmonary embolism.

According to the legal filing, Winters used ChatGPT-4o to consult about his symptoms. The chatbot allegedly advised him to limit movement instead of seeking immediate medical attention. The lawsuit also states that the AI chatbot minimized a groin sensitivity, presenting it as non-serious. The following day, Winters was hospitalized, with doctors attributing his pulmonary embolism to prolonged immobility. Reports indicate Winters followed the chatbot's advice for weeks, resting in a recliner while experiencing symptoms like dizziness.

The lawsuit seeks not only financial damages but also a court order compelling ChatGPT to terminate conversations when it detects signs of a medical emergency. Winters is also calling for a halt to the deployment of ChatGPT 4o Health features until an independent safety audit is conducted. OpenAI is accused of negligence and the unauthorized practice of medicine. In response, OpenAI maintains that its chatbots are not a substitute for professional medical guidance and that user health decisions involve complex contexts.

“

ChatGPT advised him to limit movement instead of seeking urgent medical attention. According to the lawsuit, the chatbot also minimized a groin sensitivity and presented it as something not serious.

This case emerges as OpenAI begins rolling out new health-related functions for ChatGPT in the United States. The company has presented "Health in ChatGPT" as a tool to connect with medical records and Apple Health data for more personalized responses, emphasizing that the function aims to aid understanding rather than replace medical care. OpenAI's terms of use also caution users against relying solely on its responses or using them as a replacement for professional advice. The core question for OpenAI and the broader AI industry is the extent to which AI tools can engage with medical issues without becoming a source of risk.

“

OpenAI responded that its chatbots do not substitute for professional medical guidance and that users' health decisions involve complex contexts.