AI Chatbot Prescribing Psychiatric Medication Dangers: What Utah's Experiment Reveals

# AI Chatbot Prescribing Psychiatric Medication Dangers: What Utah's Experiment Reveals

> **Quick answer:** A Utah startup called Legion Health has received state approval to let an AI chatbot renew psychiatric medication prescriptions — no doctor required per individual decision. Experts warn this creates serious patient safety risks, especially for people in crisis. The FDA has not authorized any AI system to diagnose or treat mental health conditions, and when a previous Utah AI healthcare pilot was tested with adversarial prompts, it recommended methamphetamine and suggested tripling an opioid dose. The liability question — who is responsible when the AI is wrong — remains unanswered.

An AI chatbot prescribing psychiatric medications is no longer a hypothetical. It is happening right now in Utah, and it is dividing the medical community sharply. The AI chatbot prescribing psychiatric medication dangers at the center of this debate include hallucinated drug recommendations, missed suicide risk signals, and a legal accountability vacuum that no regulator has fully closed.

## What Is Happening in Utah Right Now

In March 2026, Utah's Department of Commerce signed a one-year regulatory mitigation agreement with Legion Health, a San Francisco-based telehealth startup. Under the agreement, the company's AI chatbot can authorize refills for a defined list of non-controlled psychiatric maintenance medications — primarily SSRIs like fluoxetine (Prozac) and sertraline (Zoloft) — without a licensed physician reviewing each individual case.

The model is simple: a Utah resident pays $20 per month, describes their symptoms through the app, and the AI runs through a standardized screening. It asks about the medication's effectiveness, side effects, and flags signs of suicidality, mania, or hypomania. If the patient clears those screens, the chatbot sends the prescription to a pharmacy.