AMIE: A research AI system for diagnostic medical reasoning and conversations

AMIE is an AI system designed to engage in diagnostic conversations similar to those between clinicians and patients. AMIE is trained using a self-play based simulated learning environment and aims to improve diagnostic accuracy and conversation quality across various medical conditions and patient contexts. The study demonstrates that AMIE achieved diagnostic conversation performance at least as good as primary care physicians (PCPs), and even outperformed PCPs on multiple evaluation axes, suggesting its potential as an aid to clinicians. However, the research acknowledges limitations, and the development of AI systems aligned with the attributes of skilled clinicians is described as an inspiring challenge.