A recent study led by Harvard researchers and published in the journal Science reveals that OpenAI's AI model, 'o1 preview', has surpassed human doctors in various clinical tasks, particularly in emergency settings. Tested on 76 real cases at a Boston hospital, the AI demonstrated its ability to perform initial triage, recommend diagnostic tests, and manage patient care effectively. Independent evaluations by two doctors indicated that the AI matched or exceeded human performance at every assessment stage, especially during initial triage when limited information is available. The AI's proficiency in diagnosing rare and complex conditions, including notoriously challenging cases from Massachusetts General Hospital, further underscores its potential. Despite these impressive results, the study's authors, including senior co-author Arjun Manrai, caution against viewing AI as a replacement for doctors. They advocate for rigorous clinical trials to explore the optimal integration of AI into healthcare, emphasizing that human interaction remains crucial. Co-author Adam Rodman highlighted AI's potential in enhancing triage processes and providing reliable second opinions, with a 2025 study indicating that one in five clinicians already consult AI for such purposes.