A new article in the Journal of the American Medical Association claims that fully autonomous AI could deliver better medical outcomes than physicians. Leading experts argue the best results may come from AI alone, not hybrid teams.
A recent article in the Journal of the American Medical Association, written by Ezekiel Emanuel, Vinod Khosla, and others, argues that autonomous AI could soon deliver better medical care than even the most skilled doctors. The authors make their position clear: they believe AI, working on its own, may soon set a new standard for patient outcomes.
The JAMA article specifically highlights five key cognitive functions where AI-alone care could excel: history taking, differential diagnosis, test selection, treatment planning, and chronic disease management.
This view challenges the common belief that human judgment is still necessary for complex medical decisions. The authors point to rapid progress in machine learning, diagnostic accuracy, and data processing as signs that AI is catching up to-and may soon surpass-human doctors. They argue that as AI improves, it will be able to provide care that is more consistent, less biased, and easier to scale than what any individual physician or team can offer.
According to discussions around the JAMA article, recent studies cited show that ChatGPT o3 correctly identified the top diagnosis in 60% of complex cases, compared to just 15.9% for physicians. Another AI system achieved correct diagnoses four times more often than doctors, while using 19% fewer diagnostic tests. These findings are fueling debate about whether strict human-in-the-loop requirements might actually reduce care quality in some scenarios.
As the medical field considers the impact of fully autonomous AI, questions remain about regulation, patient trust, and the future role of clinicians. Still, the article’s authors argue that the evidence increasingly points to a future where AI alone could set the new benchmark for medical care.
