Utah AI Prescription Renewals Show Caution After Six Months

Utah breaks ground with autonomous prescription renewals

Utah became the first U.S. state in January 2026 to permit an autonomous AI system to renew prescriptions without a physician in the loop for every transaction. The pilot program, run by AI company Doctronic under an agreement with Utah's Office of Artificial Intelligence Policy, covers roughly 190 medications for chronic conditions including diabetes, high blood pressure, and depression.

Patients confirm their Utah location, answer clinical screening questions, and if cleared by the AI, receive renewals sent directly to pharmacies at $4 per prescription. Cases flagged by the system route to a physician on Doctronic's telehealth service. The arrangement sidesteps traditional medical licensing by having the state agree not to enforce unprofessional-conduct laws against Doctronic, provided the company meets contractual safety and privacy standards.

According to Forbes, which first reported the six-month data, Stanford health law expert Michelle Mello called it "one of the first deployments at scale of an autonomous, agentic system in medicine."

Early data reveal a risk-averse system

Utah's Department of Commerce released five months of pilot data in May. In 72% of cases, Doctronic's AI recommended approval and forwarded the case to a human physician for sign-off. Physicians agreed 91% of the time; the remaining 9% requested additional information such as updated lab results.

In the other 28% of cases, the AI escalated directly to a physician, flagging potential complications or the need for new testing. Physicians deemed the escalation warranted 69% of the time. In 31% of escalations, clinicians judged the system overly cautious.

Zach Boyd, head of Utah's AI office, acknowledged the system errs toward caution, which he considers appropriate for a novel deployment even as it undermines the efficiency rationale. Under the current agreement, a Utah-licensed clinician must personally review the first 250 cases before any prescription reaches a pharmacy.

Medical boards and AMA push back hard

Utah's Medical Licensing Board learned of the program through news coverage rather than official channels. In April, 11 board members sent a letter requesting suspension, citing risks of renewing medications like blood thinners without detecting changes in patient condition.

"Just because something was prescribed before does not mean it's appropriate now," board chair Dr. Alan Smith told the Associated Press. The American Medical Association echoed concerns, warning that prescription renewals are not "routine checkboxes."

Public Citizen escalated in May, backing the board's suspension call and warning that "physician supervised" pilots risk becoming rubber-stamp approvals as volume grows.

Why it matters

Utah's pilot exposes a regulatory vacuum where AI performs tasks traditionally requiring medical licenses, yet falls between state medical practice oversight and federal FDA device regulation. The program's design—using contractual agreements rather than formal licensing—creates murky accountability if AI-driven errors harm patients. Stanford's Mello warned of scope creep: "If Utah's pilot program is the proverbial camel's nose in the tent, it will be important for someone to be firmly holding the reins when the camel emerges." Texas, Wyoming, Iowa, and Idaho are already exploring similar frameworks, often using templates from the Cicero Institute, an AI-friendly think tank.

Jurisdictional questions remain unresolved

Doctronic executives told the AP they consider their tool part of state-regulated medical practice and declined to confirm whether they sought FDA clearance. An FDA spokesperson said the agency "has not authorized any AI chatbots" and is taking a hands-off approach.

University of Pennsylvania researcher Eric Bressman argued Utah accepted "the company's word on good faith" without requiring outcomes data before launch. The only published evidence on Doctronic's accuracy is a company-authored, non-peer-reviewed study showing its diagnoses matched human doctors in 80% of 500 telehealth cases.

Utah has committed to publishing findings, and Doctronic says peer-reviewed studies will arrive later this year. Whether that evidence comes before the system moves past mandatory human review will likely determine the program's trajectory and influence on other states.

These details were first reported by Forbes contributor Jesse Pines.