Six physician groups push back on claims that AI is better informed than doctors

Six of the largest physician organizations in the U.S. said Sept. 30 that claims portraying artificial intelligence (AI) as inherently better informed than physicians, or suggesting physicians can't be trusted to make clinical decisions without first consulting it, "diminish physician expertise and risk undermining patients' trust."

The joint statement came from the American Academy of Family Physicians, American Academy of Pediatrics, American College of Obstetricians and Gynecologists, American College of Physicians, American College of Surgeons and the American Medical Association (AMA). It named no one.

It arrived a day after Health and Human Services Secretary Robert F. Kennedy Jr. sat down with Vice President JD Vance at the Make America Healthy Again Summit in Washington. Kennedy told the audience AI can offer a second opinion "much better informed than any doctor in the country." He also described a meeting with OpenAI CEO Sam Altman.

According to Kennedy, Altman told him it would now be malpractice for a doctor to diagnose or prescribe "without at least checking AI."

Kennedy grounded his case in time and volume. He said a patient's record can run a thousand pages, that a physician with six minutes for a visit can't review it and that AI can. Earlier in the summit, Felipe Millon, head of OpenAI's government go-to-market division, told attendees, "It is medical malpractice not to get a second opinion from AI today." AI companies including OpenAI and Anthropic helped sponsor the event, The Epoch Times reported.

What did the physician groups say?

"Physicians evaluate patients in context, drawing on years of training and experience," the statement reads. "They ask questions, understand each patient's unique needs and circumstances, exercise professional judgment and take responsibility for the care they provide."

"Physicians have long embraced technologies that improve care, and AI has tremendous potential to provide new tools and insights," the groups wrote. "But patient safety, physician expertise and the humanity of clinical practice must guide how these tools are developed and used."

"Pitting physicians against technology or suggesting AI should replace physician expertise does not advance this important work," the statement continues. "We need to move carefully, and we need to get this right."

Earlier in September, a paper in JAMA by bioethicist Ezekiel Emanuel and venture capitalist Vinod Khosla concluded that for cognitive medical functions, AI-alone care "is likely to be better than physician-only or physician-AI hybrid care."

AMA CEO John Whyte criticized the paper in Wired, noting that some of the studies it surveyed were simulations rather than blinded experiments and saying the AMA sees potential in the tools only when they are used within a care plan governed by a physician.

In an AMA survey of nearly 1,700 physicians released this year, 81% reported using AI professionally, more than double the 2023 rate. In the same survey, 85% said they want a say in AI adoption in their practice, and respondents named clear liability frameworks as a top regulatory priority.

Would skipping AI really be malpractice?

"If AI produces a recommendation that's different than the standard of care, and the doctor follows it, and the outcome is actually adverse, well, then by definition, the doctor has violated the standard of care," Richard E. Anderson, M.D., FACP, chairman and CEO of The Doctors Company and TDC Group, said in February on the Off the Chart podcast.

When AI departs from the standard of care and the patient does well, there is no damage and no lawsuit, Anderson said. He expects that imbalance to slow clinical adoption, because "using it for recommendations that are different than what your clinical judgment would suggest actually exposes you to liability."

"Eventually, AI will become the standard of care," Anderson said, though he predicted the path there would be "piecemeal," varying by venue and by system. "I'm certainly not suggesting that we don't use it or that it can't be a benefit, but I am saying that putting it into the infinitely complex world of health care organization in the United States is a heavy lift."

Marc Succi, M.D., associate professor of radiology at Harvard Medical School and executive director of the Mass General Brigham MESH Incubator, sorts AI uses by risk.

"I would say comfortable using AI for low risk, high feasibility tasks," he said in an April interview, listing ambient documentation, visit notes, summarization, patient-friendly explanations and billing. "Things that if you get it wrong, okay, it's not great. You lose some money, but a patient doesn't suffer."

Clinical decision support, replies to patient messages, lab orders and medication renewals belong in a different category, Succi said. "That's where you got to stop and look at the level of performance very critically in multiple different ways, not just trust what the vendors say."

"At the end of the day, it's how much liability risk are you willing to take as a doctor?" Succi said.

Where does AI fit in the exam room today?

Nancy Cibotti, M.D., a primary care physician and the U.S. chief medical information officer at AI scribe company Heidi, described her own routine during an April Medical Economics and Physicians Practice webinar that Heidi sponsored.

She uses a scribe to capture the visit and an evidence tool to check guidelines. "But then ultimately it's up to me and the patient to figure out what really applies to them," she said. "There are things that the AI tool may not be able to do. They don't know the patient."

An informal flash poll of 103 physicians, fielded for that webinar and not scientific, split 55% to 45% on whether AI is making medicine better. Respondents' two leading worries were AI replacing clinical judgment and AI disrupting the physician-patient relationship.

Josh Umbehr, M.D., co-founder of direct primary care practice Atlas MD, called himself a technophile on a July Off the Chart episode. "At its core, it's decreasing the work of work," he said. "Note taking doesn't add value. The hour-long conversation with the patient, where you're keyed in and not typing and not trying to juggle what to remember and what not to, that's high value."

Umbehr also said the current system "pits AI against the doctors." Faster charting gets converted into fuller schedules, he said, and insurers use AI of their own to counter the notes physicians' AI tools produce.

Who decides how the tools are used?

David Carmouche, M.D., chief medical and commercial officer at value-based care company Lumeris, was appointed this year to a federal Healthcare Advisory Committee that advises Kennedy and Centers for Medicare & Medicaid Services Administrator Mehmet Oz.

"I think health care has benefited from slow adoption at times," Carmouche said on a May Off the Chart episode. "There's a safety reality. This is a new, powerful technology."

"It's okay to be skeptical, but I think being skeptical without information and without understanding and learning is challenging," he said. "We should all, as physicians, shape how this technology is going to be deployed."

The six organizations ended their statement on the same point: "As medical societies representing physicians across the country, we are committed to ensuring physicians help lead the responsible development and use of AI in medicine."