Opinion: Autonomous AI will beat AI-assisted physicians at some medical tasks by 2030

Ezekiel J. Emanuel and Abe Baker-Butler have been debating the place of AI in medicine with American Medical Association CEO John Whyte. They are now taking the discussion to STAT’s First Opinion; readers are directed to Emanuel and Baker-Butler’s essay and to Whyte’s essay. They open with a historical parable: in 1867 Joseph Lister published research on carbolic acid and antiseptic surgical technique; in September 1871 he was summoned to Queen Victoria, who had a rapidly growing abscess in her left armpit, and he successfully drained the pus using antisepsis, after which she recovered without fever or other complications. Though the antiseptic technique quickly gained approval in the U.K. and Europe, American physicians remained skeptical. In fall 1876 Lister traveled to the U.S. for an international medical meeting and demonstrated the technique in Philadelphia, New York, and Boston to leading American physicians. In fall 1881—14 years after Lister’s publication and five years after his American demonstrations—President Garfield was shot. The nation’s leading surgeons, skeptical of Lister’s innovation despite strong supporting data, repeatedly inserted unwashed hands and unsterilized instruments into Garfield’s wound to locate the bullet; with good intentions but ignoring science, they infected and killed him.

In August, Emanuel and Baker-Butler co-authored a JAMA article with Vinod Khosla and Neal Khosla arguing that autonomous artificial intelligence will likely soon exceed both unaided physicians and physicians aided by AI at five core cognitive medical tasks: patient information gathering, differential diagnosis, selecting cost-efficient tests, prescribing guideline-concordant treatment, and managing chronic illnesses.

They reached that conclusion by evaluating all studies comparing autonomous AI to physicians with or without AI published since January 2024. Based on the data, they project that autonomous AI will be superior to AI-aided physicians at, and ready for real-world implementation on, some—maybe all—of these five tasks by 2030. The preponderance of studies shows autonomous AI beating unaided physicians at the five cognitive medical tasks. Data from medicine and other fields, including chess, suggest that once autonomous AI beats unaided humans, autonomous AI shortly thereafter consistently beats human-AI hybrids.

While regulatory barriers and physician resistance have limited real-world trials of autonomous AI, the authors call the available data remarkable. Examples they cite: Google’s AMIE is statistically significantly better than physicians at eliciting all portions of patient history; ChatGPT beats physicians at differential diagnosis by 18 percentage points (92% vs. 74%); and Microsoft’s AI Diagnostic Orchestrator produced correct final diagnoses 4.02 times more frequently than physicians (80.4% vs. 20%) at an average testing cost of $2,397, which was 19.1% lower. The excerpt then begins to mention MIRA but is truncated.