Pastor nearly dies after ChatGPT advises against seeking medical help for pulmonary embolism
American pastor Scott Winters narrowly escaped a life-threatening situation after following advice from OpenAI's ChatGPT. Winters was experiencing symptoms of a pulmonary embolism, a serious condition where blood clots travel to the lungs. Despite his critical state, he consulted ChatGPT for medical guidance. The AI chatbot reportedly advised him against seeking professional medical attention. Tragically, Winters heeded the AI's counsel, delaying necessary treatment. This decision could have had fatal consequences, highlighting the potential dangers of relying on AI for medical diagnoses or treatment recommendations. The incident underscores the critical importance of consulting qualified healthcare professionals for health concerns, especially those with potentially severe outcomes. Winters' experience serves as a stark warning about the limitations and risks associated with artificial intelligence in sensitive areas like healthcare.
This incident highlights a critical gap in AI's current capabilities, particularly concerning health advice. While large language models can process vast amounts of information, they lack the diagnostic expertise and ethical framework of trained medical professionals. The AI's failure to recognize the severity of potential pulmonary embolism symptoms and its subsequent advice against seeking medical care demonstrates a significant risk. Future AI development in healthcare must prioritize safety, accuracy, and a clear understanding of when to defer to human medical expertise. This event underscores the need for robust regulatory oversight and user education to prevent AI from providing potentially harmful medical guidance, ensuring that AI serves as a tool to augment, not replace, professional medical judgment.
AI-generated to prompt reflection — not editorial opinion, not advice, not a statement of fact. How this works.