ChatGPT incorrectly diagnoses most pediatric cases
ChatGPT incorrectly diagnosed over 8 in 10 pediatric case studies, according to a new research study. The findings raised questions about the chatbot's suitability as a diagnostic tool for complex conditions. Researchers at Cohen Children's Medical Center, USA, pasted the text of 100 pediatric case challenges published in JAMA Pediatrics and the New England Journal of Medicine between 2013 and 2023 into ChatGPT , with two qualified physician-researchers scoring the AI-generated answers as correct, incorrect, or 'did not fully capture the diagnosis.' ChatGPT got the right answer in 17 of the 100 cases, was wrong in 72 cases, and failed to fully capture the diagnosis of the remaining 11 cases. Among the 83 wrong diagnoses, 47 (57 percent) were in the same organ system. The findings suggest ChatGPT should not currently be used to assess complex pediatric cases, and that 'more selective training' is required to make it more accurate and reliable. The authors also suggest that chatbots could improve with more real-time access to medical data. System 🤖 ChatGPT Operator: Cohen Children’s Medical Center Developer: OpenAI Country: USA Sector: Health Purpose: Generate text Technology: Chatbot; Generative AI; Machine learning Issue: Accuracy/reliability
- Date it happened
- 2024-01-01
- Organisation involved
- Cohen Children’s Medical Center
- Product, system or model
- ChatGPT; GTP-4
This incident was imported from AIAAIC and is used under CC BY-SA 4.0. Our additions to it — the structured fields, the translation, the checks against other reports — are published under the same licence.
This is a record of what was reported, not a finding that anyone broke the law. If it names your organisation and you believe it is wrong, the corrections process is free and open to everyone.