Financetoday
financialTelegram

Medicare's AI Review Pilot Coincides With a Sharp Drop in Approved Care

Health Policy·October 29, 2026

Medicare's AI Review Pilot Coincides With a Sharp Drop in Approved Care

Medicare is testing whether artificial intelligence can speed up prior authorization, the review process that decides whether a recommended treatment or service gets approved before it is paid for. The stated aim is to target wasteful medical spending. The reported outcome has drawn attention for a different reason: once the automated reviews started, the volume of approved services fell sharply.

Prior authorization has long been a source of friction in the U.S. health system. Physicians often wait days for decisions, and patient advocates argue that the process delays care and produces denials that many patients never appeal. Using AI to screen requests is meant to clear routine, low-risk claims quickly and flag the ones that look unnecessary or unusual for closer human review.

The drop in approvals is what makes the pilot stand out. It suggests the automated system is rejecting or holding back more requests than the earlier process did. Whether that reflects fewer unneeded procedures or patients going without care they actually required is the central question, and the headline and summary alone do not answer it.

That uncertainty sits at the heart of the debate. Supporters of tighter review argue that Medicare has a duty to avoid paying for low-value care, and that software can apply coverage rules more consistently than overloaded reviewers. Critics counter that automated systems can be hard to audit, may lean toward denial, and could hit patients with complex or rare conditions the hardest.

For providers and health finance watchers, the pilot is a signal of where Medicare cost control may be heading. If AI-assisted review can trim spending without cutting necessary care, expect it to expand. If the reduction falls on patients who needed treatment, expect pushback from doctors, hospitals and lawmakers. Upcoming utilization data and any disclosure about how denials are appealed will show which outcome is taking shape.

Reporting based on an external source.