AI-Only Care Projected to Surpass Physicians, AMA Raises Concerns

A paper in the *Journal of the American Medical Association* surveyed hundreds of AI-medicine studies and concluded that AI already exceeds physicians on five core tasks—eliciting patient information, diagnosing, ordering tests, and managing chronic disease. They further predict that, within four years, human-in-the-loop models could actually degrade performance compared with AI operating independently.

Futurism

John Whyte, CEO of the American Medical Association, disputes the paper’s methodology, noting that many cited studies are simulations rather than blinded trials, the gold standard for clinical evidence. He also points to a *Nature* article that found most patients struggle to obtain expertise from AI chatbots. Whyte emphasizes that while the AMA acknowledges AI’s potential, tools must operate within physician-governed care plans.

Futurism

Internal data from OpenEvidence, an AI chatbot built for clinicians, indicate that roughly two-thirds of U.S. physicians are already using the tool. In the United Kingdom, a team of neurosurgeons recently employed an AI system to guide tumor-removal surgery. He likens fears of doctor displacement to the “doorman fallacy,” arguing that highly trained professionals remain indispensable.

Futurism

• “The AMA does see the potential in these tools, but they have to be utilized in the context of a care plan that’s governed by a physician,” — John Whyte

Futurism

• “We’re making a prediction for four years from now,” — Ezekiel Emanuel

Futurism