OpenAI o1은 응급실 환자의 67%를 정확히 진단했고 분류 의사는 50~55%를 기록

OpenAI o1's superior performance over human doctors in ER triage marks a significant shift from AI passing exams to providing practical clinical reasoning in high-pressure environments. This highlights AI's growing role as a reliable second-opinion tool that could reshape the medical industry into a collaborative tripartite model between doctors, patients, and AI.