TL; DR
Utah has become the first US state to allow its AI chatbot, Doctronic, to renew prescriptions without a doctor through a regulatory sandbox that waives licensing laws. The state’s medical licensing board, blindsided by the January release, requested in April that the pilot be halted because of safety risks, but the state refused. The case exposes a federal-state regulatory vacuum around artificial intelligence in medicine.
Utah has quietly become the first US state to allow an AI chatbot to renew prescriptions without a doctor. According to the Associated Press. The program, run by a company called Doctronic, launched in January and sparked fierce medical debate.
Residents can skip the doctor’s office and fill prescriptions online via a chatbot. It asks about their medications and history, checks the national pharmacy database, and updates the script or forwards it to the person’s doctor.
The launch was made possible only through a “regulatory sandbox” This allows Utah officials to waive laws promising AI. State and federal regulations restrict prescribing to otherwise licensed medical professionals.
Dr. Eric Bressman of the University of Pennsylvania told the AP: “We’ve crossed the line in terms of giving a medical license to something that’s not human, whether we want to call it that or not.” He and others say they’re not opposed to AI prescribing, but want it to be as strict as the standards for human doctors.
The board left on the edge
Utah’s medical licensing board says it only learned of the program when news broke of the January launch. 11 members in the April letter called the pilot to stopciting the risks of auto-renewable drugs with side effects or interactions.
“We were essentially told, ‘Yes, it’s going ahead.’ No, you don’t have a say in this,” said Dr. Alan Smith, a family physician who chairs the board but spoke for himself.
state refused to stopnoting that human doctors still review each refill at this early stage.
The program is currently overseen by a five-member board of AI experts, none of whom are physicians. Doctronic expects to move to a fully automated filling system soon.
Smith cautions that the risks are real, noting that about 190 of Doctronic’s refillable drugs contain blood thinners, which can be dangerous if a patient experiences internal bleeding. The American Medical Association reiterated its concern that “prescription renewals are not simple check boxes.”
Regulator vacuum by design
This case reveals a jurisdictional tangle because medical technology is regulated at the federal level, while medical professionals are regulated by the states. Doctronic touts its AI as part of state-regulated medical practice, though some experts argue that it has crossed into FDA territory.
The company would not say whether it has received FDA approval. The agency told the AP that it does not authorize any AI chatbots, but wants to encourage innovation. Broader easing of controls on AI health tools.
Critics see the history as rhyming, Bressman compares the moment to casual medicine at the turn of the 20th century, when plates and criteria did not exist. The template for licensing AI medical services in other states comes from the pro-AI Cicero Institute, founded by Palantir co-founder Joe Lonsdale.
As security researchers warn, the stakes are not abstract medical chatbots can sound authoritative while offering dangerous advice. Others focus on it turning people away from care can undermine its promised outcomes.
Competitors are also trying to map these failure modes. It went all the way to Meta posing as teenagers to test how competing chatbots handle sensitive topics.
Doctronic plans peer-reviewed studies later this year, although its only published paper to date was written by its own scientists and has not been independently peer-reviewed. As one Utah law professor put it, companies risk allowing the technology race beyond the evidenceand betray public trust in the process.
