Is ChatGPT a Reliable Auxiliary Tool in Basic Life Support Training and Education? A Cross-sectional Study.

Background and aims: The Basic Life Support (BLS) training forms an integral part of the medical education. Affordable and easily accessible auxiliary tools can aid in delivering this training to medical professionals. This cross-sectional observational study explored the utility of ChatGPT in BLS training and education. Methods: Two authors prompted 30 subjective questions based on pediatric and adult BLS on ChatGPT-4.0 10 times each, generating 600 responses. Responses were rated for accuracy using a five-point Likert scale, and repeatability was analyzed using concordance metrics, simulating varied user interactions with the chatbot. Statistical analysis was done using STATA version BE 18 software. Results: The accuracy of ChatGPT-4.0 to BLS-related responses was 48.67%, with a 95% confidence interval between 44.67 and 52.67%. The findings of the study demonstrated moderate repeatability of ChatGPT. Conclusion: ChatGPT-4.0 demonstrated moderate repeatability and limited accuracy, suggesting that while it may serve as an adjunctive self-learning tool, it cannot currently replace traditional instructor-led BLS training. Professionals should critically review and verify the ChatGPT-4.0 content before using it in BLS. How to cite this article: Gupta P, Pandey AK, Kumar A. Is ChatGPT a Reliable Auxiliary Tool in Basic Life Support Training and Education? A Cross-sectional Study. Indian J Crit Care Med 2025;29(8):684-691.