A study comparing an AI model’s clinical reasoning with that of physicians found the AI outperformed residents and attending physicians in simulated cases, despite more instances of incorrect reasoning. The study involved 39 physicians and the AI model GPT-4 evaluating 20 clinical cases. The AI achieved a higher clinical reasoning score compared to physicians and had a higher probability of scoring well. While the AI had more incorrect reasoning instances, it performed similarly to physicians in diagnostic accuracy. The study suggests that AI may augment rather than replace human reasoning, with potential for improvement through further training. The research was published in JAMA Internal Medicine on April 1, 2024.
Source link
AI Surpasses Harvard Docs on Clinical Reasoning Test
