Utah Medical Board Demands AI Prescription Pilot Suspension

The Utah Medical Licensing Board has formally requested immediate suspension of a first-in-the-nation pilot program that allows an artificial intelligence system to approve prescription medication renewals. The board sent a letter to the Utah Department of Commerce arguing the program was implemented without proper medical oversight, potentially placing citizens at risk.

At the center of the controversy is Doctronic, an AI health platform vendor contracted by the state to process 30-, 60-, or 90-day renewals for medications already prescribed by licensed physicians. The pilot launched in October 2025, with official announcement following on January 6, 2026. Board members argue they should have been included in the process before the program's implementation, not once the system was already live.

"Proceeding with this agreement without consulting the Medical Board potentially places Utah citizens at risk and remains a major concern of the board," the letter states. "It is imperative that professionals with medical backgrounds review all proposals prior to implementation to ensure these programs do not compromise patient safety."

The board's concerns are grounded in clinical reality. Each refill requires reassessment to determine safe dose adjustments, side effects monitoring, contraindications, or new drug interactions. Patients who continue refilling medications without assessment may remain on outdated or suboptimal therapy for months or years. There's a reason prescription refills require physician authorization (and it's not just bureaucratic red tape).

According to the Utah News Dispatch, the program covers 192 drugs and requires consumers to request an AI-generated refill through the Doctronic website. The state's Office of Artificial Intelligence Policy says the system operates within strict parameters and with physician oversight, designed to lighten clinicians' workload, improve refill continuity, and expand access for patients.

Currently, the pilot is in its first phase, which requires a licensed medical practitioner's approval before issuing a renewal. The state hasn't received any serious safety incidents. However, a third-party red teaming report identified potential vulnerabilities under adversarial usage. This is the kind of detail that matters when you're talking about medication safety.

Independent reporting from MedPage Today confirms the board was not notified about the pilot program until after it had gone into effect. The Utah Department of Commerce told KUTV that the board was not required to be involved and said licensed medical professionals were consulted before launch. Officials also said that all prescriptions are reviewed by a licensed physician and that there is no autonomous prescribing at this stage.

A spokesperson for the Department of Commerce said in an email that the pilot program is in Phase One, which requires the first 250 prescriptions of each medication class to be reviewed by a physician to assess the accuracy of the AI. In future phases, it is expected that the AI will work autonomously, with some level of physician interaction or review. This progression is where the real concern lies.

Matt Pavelle, co-founder and co-CEO at Doctronic, said in a statement the company is participating in the process as designed, with designated safeguards and physician participation. "We remain focused on demonstrating safe, evidence-based expansion of routine care access," he wrote. Each phase of the tool deployment has extensive identity and prescription verification protocols, and will be supervised by physicians.

The system won't be able to issue new prescriptions, handle controlled or addictive substances, or make changes to treatment plans. These guardrails matter, but they also raise questions about where the line gets drawn when the program expands.

In a response to the board's letter, the Utah Division of Professional Licensing and the Office of Artificial Intelligence Policy said the pilot won't be suspended at this time, since it's still in phase one, which requires physician reviews for each refill. However, the Office of Artificial Intelligence Policy may modify or cancel the program if safety benchmarks aren't met.

The offices added that any agreement involving healthcare decisions are also made in consultation with medical specialists in that specific field. "The pilot involving Doctronic was rigorously reviewed by several medical professionals prior to launch. This evaluation process generated a large number of suggested substantive adjustments and guardrails, many of which were integrated into the pilot," the offices wrote.

However, the offices committed to working in collaboration with the board in future steps of the pilot and to vet future healthcare proposals. They also contested the "financial motivations" the board referenced in its letter. "Neither the State of Utah nor the Utah Department of Commerce has any motivation involved in the specific outcome of this policy project other than determining the best public policy for Utahns," the offices wrote.

Not everyone is opposed to the concept. Michelle Mello, PhD, of Stanford Law School, wrote in JAMA Health Forum that review of Doctronic's proposal reveals a well-designed pilot program with several guardrails against patient harm. Only adults can participate and the selected medications are established therapies for common health conditions such as hypertension, depression, birth control, diabetes, glaucoma, asthma, inflammation, and hyperlipidemia.

Mello did, however, express some concerns about scope creep. "Once an AI system has secured acceptance, vendors may be able to push updates that include substantial changes without attracting the same degree of scrutiny as the initial adoption," she noted. "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."

Michelle McOmber, CEO of the Utah Medical Association, told MedPage Today the program is "completely inappropriate for a number of reasons." For instance, the patient may be on other medications that will interact with a refill, and there are concerns related to HIPAA and privacy concerns since there is nothing that says these requests and the records they generate are secure.

"If it is a refill, whose name is it under and who has liability?" McOmber said. If the prescription renewal is listed under the original physician's name, that is a problem when the physician is not involved in the refill. That person would be liable if anything goes wrong. In addition, it may no longer be an appropriate prescription since a patient's health may have changed.

The broader context matters here. Technology is being used to expand healthcare access for rural and uninsured populations in many ways, especially in states like Utah that have an AI sandbox program that is attractive to health AI developers like Doctronic. Last year, a federal AI Action Plan called for establishing regulatory sandboxes or AI Centers of Excellence to deploy and test AI tools.

Earlier this month, Doctronic, which raised $40 million in Series B funding, was announced as one of 150 accepted applicants under the Centers for Medicare & Medicaid Services' Advancing Chronic Care with Effective, Scalable Solutions program, a results-based Medicare payment model. The company has agreed to participate in the voluntary reimbursement program under the Early Cardio-Kidney-Metabolic track.

Whether this pilot succeeds or fails, the real question is whether patients will actually benefit or whether this becomes another case of technology moving faster than regulation can keep up. The board's concerns about patient safety are legitimate, but the state's argument about expanding access to care for underserved populations also has merit. Someone's going to have to make the call on which priority wins out.

Artūras Malašauskas is an AI Systems Integrator with 20+ years of production-grade web engineering experience. He has designed, shipped, and scaled enterprise Python/PHP systems for logistics, SaaS, and public-sector clients. For the past year, he has focused exclusively on AI integrations: deploying open-source LLMs, building generative media pipelines (image, audio, video), and engineering multi-agent workflows for real production environments. His standard: reproducibility, security, cost-efficient inference—no vaporware. He documents and evaluates emerging AI tooling, separating verified capabilities from marketing noise. Technical editor at: muza-ai.eu, ai-verslas.lt, ai-naujinos.lt Connect on LinkedIn