The format of AI explanations fundamentally shapes physician decision-making. A randomized controlled trial demonstrates that chain-of-thought reasoning from GPT-4 improves radiologist diagnostic accuracy by 12.2% compared to standard AI output. Original paper: The effect of medical explanations from large language models on diagnostic accuracy in radiology. — NPJ digital medicine. 10.1038/s41746-026-00001-1 📄 Read the article