
The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
The Centers for Medicare & Medicaid Services launched an AI-driven prior authorization pilot in six states, enlisting six vendors. Early issues included technical failures, slow determinations, blank denial letters, and missing patient portals, leading to delayed or denied medical care and reduced service utilization, prompting public backlash and political scrutiny.[AI generated]
Why's our monitor labelling this an incident or hazard?
The AI system is explicitly involved in prior authorization decisions affecting Medicare services, directly influencing patient care by approving or denying procedures. The reported delays, technical failures, and automation bias indicate malfunction and misuse risks. The harm includes potential denial or delay of necessary medical care (harm to health), increased administrative burdens (disruption), and concerns about transparency and rights violations. The public backlash, legal challenges, and political responses further confirm the significance of the harm. Hence, this event meets the criteria for an AI Incident rather than a hazard or complementary information.[AI generated]