Utah to let AI examine patients and prescribe medication without human oversight
Utah has approved a pilot program allowing an AI system to diagnose mild-to-moderate acne from photos and prescribe a limited set of topical medications without real-time human oversight. Commenters debate whether this is a sensible way to reduce low-value doctor visits and loosen overly strict prescription rules, or a risky precedent that shifts power to profit-driven companies, weakens accountability, and expands health-data collection. Many note the program’s narrow scope and phased doctor review, but see it as the “thin end of the wedge” for broader AI-driven medical care.
Scope of the Utah Pilot
- Program is a one‑year, narrowly scoped pilot for mild‑to‑moderate acne.
- Patients use a $4.99/month app, submit facial photos and medical history; AI analyzes skin and prescribes from a short list of topical medications.
- Rollout is staged: first 100 cases fully reviewed before filling; next 400 reviewed retrospectively; later only ~10% audited monthly.
- System explicitly excludes higher‑risk drugs like isotretinoin/Accutane and oral/systemic meds.
Arguments That It’s Beneficial
- Frees patients from “time‑wasting” visits and long waits for simple issues like acne, UTIs, chronic minor conditions.
- Similar or lower risk than current minimal telehealth or “text‑only” prescribing flows.
- Seen as a pragmatic way to expand access where doctor appointments are months out.
- Viewed by some as a low‑stakes testbed to mature AI tools before tackling harder problems.
- Some argue that if nurses, PAs, or pharmacists can safely handle such cases, AI can too.
Concerns and Criticisms
- Slippery‑slope fears: “narrow end of the wedge” toward broader, riskier AI prescribing or “pill mills.”
- Worry that patients “who know what they need” are often confidently wrong, masking serious conditions.
- Objection that if drugs are safe enough for AI, perhaps they should just be OTC instead.
Prescription vs OTC and Regulation
- Several included medications are already OTC in weaker forms; prescriptions largely gate higher strengths or riskier usage.
- Some see US prescription rules as over‑paternalistic compared to other countries; AI workarounds highlight a broken system rather than fix it.
Data Privacy and Commercial Incentives
- Strong concern about uploading face images and PII to a startup, given telehealth’s record of data sharing with ad platforms.
- Subscription model plus data capture seen by some as adding costs, tracking, and lock‑in rather than genuine “liberalization.”
Ethics, Accountability, and Risk
- Debate over who is accountable when AI harms patients: company, overseeing doctors, or end users via click‑through ToS.
- Some argue harm is likely small given vetted topicals; others say dismissing risk is irresponsible.
- A minority stance is that AI in direct medical care should be banned; others see careful, audited deployment as acceptable.