AI Just Got a Medical License in Utah—And Doctors Are Sounding the Alarm

Utah has quietly crossed a threshold that many thought was years away: granting an artificial intelligence system the authority to renew prescriptions without mandatory physician oversight. Since January 2026, the state has allowed Doctronic's AI platform to autonomously process refills for approximately 190 chronic disease medications—and the five-month pilot data released in May reveals both the promise and peril of this unprecedented experiment.

The numbers tell a complex story. The AI recommended approval in 72% of cases, with physicians agreeing 91% of the time when they reviewed those decisions. When the system flagged concerns and escalated cases to human clinicians—which happened 28% of the time—doctors agreed with the escalation 69% of the time. In 31% of escalations, physicians judged the AI to be overly cautious, suggesting the technology may err on the side of safety.

But these statistics have done little to calm growing opposition from medical professionals, ethicists, and regulatory bodies who argue Utah has opened a Pandora's box.

A Regulatory End-Run Around Traditional Oversight

Utah's approach bypassed traditional FDA clearance pathways entirely. Instead, the state's Office of Artificial Intelligence Policy struck a deal with Doctronic: the company would adhere to contractual safety guardrails, and in exchange, the state wouldn't enforce unprofessional-conduct laws against the AI system's autonomous decisions.

The arrangement covers 30-, 60-, and 90-day renewals for medications treating chronic conditions like hypertension, diabetes, and depression, at $4 per prescription. Controlled substances, new prescriptions, and treatment plan changes remain off-limits. The first 250 cases required personal review by a Utah-licensed clinician before pharmacy fulfillment.

"Just because something was prescribed before does not mean it's appropriate now," warned Dr. Alan Smith, chair of Utah's Medical Licensing Board, which signed an April suspension request with 11 members opposing the program.

The Accountability Gap That Has Experts Worried

Stanford bioethicist Michelle Mello described the initiative as "one of the first deployments at scale of an autonomous, agentic system in medicine" and identified two critical structural risks: unclear liability and accountability frameworks, and the potential for scope creep beyond the initial narrow use case.

University of Pennsylvania's Eric Bressman put it more bluntly: "We have crossed a threshold in terms of giving something that is not human a medical license, whether or not we want to call it that."

The evidence base supporting Doctronic's system remains thin. The only published accuracy data comes from a company-authored, non-peer-reviewed study showing 80% diagnostic concordance with human doctors in 500 telehealth consultations. The FDA has stated it "has not authorized any AI chatbots" and is taking a hands-off approach, while Doctronic declined to confirm whether it even sought FDA clearance.

A Model Poised for National Expansion

Despite the controversy, Utah's experiment is already spawning imitators. The state signed a second agreement in March 2026 with Legion Health for psychiatric maintenance drugs, with stricter guardrails requiring 98% concordance for the first 250 cases and 99% for the next 1,000.

Texas and Wyoming are reportedly developing similar AI carve-outs, while Iowa and Idaho have introduced AI medical service licensing bills based on templates from the Cicero Institute—a think tank founded by Palantir co-founder Joe Lonsdale. The institute's health policy director, Adam Meier, frames physician resistance as economic protectionism rather than legitimate safety concerns.

For patients, the program offers faster, cheaper prescription renewals and reduced access barriers for chronic disease management. But the 28% escalation rate reveals that even "routine" renewals carry meaningful clinical complexity that automation may miss—validating concerns that AI could overlook condition changes requiring physician assessment.

As other states watch Utah's experiment unfold, the fundamental question remains unanswered: Has the state found an innovative pathway to expand healthcare access, or has it created a regulatory loophole that sidesteps essential safeguards? The next six months of data may determine whether this model scales nationally or triggers the regulatory backlash that critics predict.

Sources

1. Utah's Letting AI Renew Prescriptions. Here's The Surprising Result — Forbes