Relying on AI for health advice? Study finds 80% early diagnoses go wrong
This image was generated by AI and may not depict real events.
A new study found that AI chatbots misdiagnose medical conditions in over 80% of early clinical cases, highlighting serious gaps in AI-led diagnosis. The research assessed 21 large language models across 29 clinical case scenarios, generating a total of 16,254 diagnostic responses.
Researchers found that AI chatbots frequently get health diagnoses wrong, especially in early stages. The study evaluated 21 large language models, including systems from OpenAI and Google. AI systems failed to provide the correct diagnosis in over 80% of early-stage cases. Failure rates fell to less than 40% for final diagnoses with more complete data. The study's lead author noted that AI models struggle at the open-ended start of a case, when there isn’t much information. Experts say the issue lies in how these systems generate responses, predicting patterns based on vast datasets rather than reasoning clinically.
This content was automatically generated and/or translated by AI. It may contain inaccuracies. Please refer to the original sources for verification.