Utah could become the first U.S. state to allow artificial intelligence systems to examine patients and prescribe medication without direct human oversight. The proposed approach would give AI a far more independent role in healthcare, moving beyond administrative support or assisting physicians with routine decisions. Instead, an AI system could potentially evaluate a patient’s condition and determine an appropriate prescription on its own. Supporters may view the policy as a way to expand access to care and reduce pressure on medical professionals, particularly in underserved areas. Critics, however, are likely to question whether current AI systems can reliably handle complex diagnoses, account for individual medical histories, and respond safely when symptoms are ambiguous. The plan also raises broader concerns about accountability: if an AI makes a harmful decision, it may be unclear who bears responsibility. Utah’s move could become an important test of how regulators balance innovation, accessibility, patient safety, and human supervision in AI-powered healthcare.