- AI Healthcare Insider
- Posts
- Medicare's AI gatekeepers are here, and they're already making decisions about your care
Medicare's AI gatekeepers are here, and they're already making decisions about your care
Also, Utah lets robots refill your prescriptions, Big Tech's healthcare AI raises red flags, and the unglamorous AI use case that's actually working.
Ram Menon
February 09, 2026 • Estimated Reading Time: 3 minutes
🌟 Editor's Note
This week marks a pivotal moment in healthcare AI: the federal government is testing AI for Medicare claims processing, while Utah becomes the first state to let AI handle prescription refills autonomously. Meanwhile, as ChatGPT and Claude expand their healthcare footprints, industry leaders are drawing important lines about where AI belongs and where it doesn't. Plus, we spotlight how intelligent appointment automation is becoming healthcare's quiet AI success story.
CMS launches AI-powered prior authorization pilot in 6 states
The Centers for Medicare & Medicaid Services has officially launched the Wasteful and Inappropriate Service Reduction (WISeR) Model, marking the first time AI and machine learning will assess prior authorization requests for Original Medicare. The six-year pilot operates in Arizona, New Jersey, Ohio, Oklahoma, Texas, and Washington, affecting 6.4 million Americans.
The program targets services CMS identifies as vulnerable to waste and abuse, including skin substitutes, nerve stimulator implants, and knee arthroscopy for osteoarthritis. Technology companies will use AI alongside human clinical review to process requests within 72 hours.
CMS emphasizes that licensed clinicians, not machines, make final decisions. However, a 2024 Senate committee report found AI tools linked to care denial rates 16X higher than human-only decisions
The boring AI use case that's quietly making hospitals millions: appointment scheduling
While debates swirl around AI in clinical decision-making, one use case is quietly proving its value: intelligent appointment scheduling.
Connie Barrera, CISO at Jackson Health System
Connie Barrera, CISO at Jackson Health System, calls it her organization's most successful AI implementation. "When patients reschedule, our AI ensures all associated orders, referrals and care team notifications remain coordinated, eliminating manual reconciliation that previously consumed significant staff time while creating opportunities for clinical errors."
The Results?
No-shows: way down
Appointment availability: way up
Staff time wasted on phone tag: practically eliminated
Another top-ranked health system partnered with Avaamo to deploy AI agents across web, portals, IVR, SMS, and voice. The patient scheduling agent fully automated 81% of appointment confirmations with over a million scheduling interactions handled by AI each year. And only 2% of patients even ask for a human.
This exemplifies a broader pattern: AI's greatest near-term value may come from removing administrative "pebbles" rather than making clinical decisions.
Utah just became the first state to allow AI prescription refills
Utah has made history as the first state to let AI legally participate in prescription renewals. Through a partnership with Doctronic, patients with chronic conditions can now renew 190 commonly prescribed medications without speaking to their doctor's office for $4 per refill.
The AI verifies identity, reviews medical history, checks for drug interactions, and sends approved prescriptions directly to pharmacies. Higher-risk medications including painkillers and ADHD drugs are excluded, and flagged cases escalate to human physicians. Doctronic reports its AI matched physician recommendations in 99.2% of cases. The company's co-founder expects "a dozen other states" to approve similar programs in 2026, though the AMA has urged caution about AI operating without physician input.
ChatGPT for healthcare and Claude AI pose governance challenges
With OpenAI's ChatGPT for Healthcare and Anthropic's Claude for Healthcare now live, major health systems including Cedars-Sinai, HCA, Memorial Sloan Kettering, and Stanford Medicine are integrating LLMs into patient care. Elation Health reports Doctors using Claude are getting answers 61% faster.
But healthcare leaders are drawing a clear line. "The real governance challenge is the decoupling of certainty from accountability," noted Adam de la Zerda, CEO of Visby Medical.
The consensus emerging: when used thoughtfully, LLMs like GPT and Claude belong before and after the point of care, "not autonomously in it."
What this means for you
Healthcare AI is advancing on two tracks: autonomous clinical decisions (Utah prescriptions, WISeR claims) with significant guardrails and debate, and administrative optimization with measurable ROI and clinician acceptance.
If you're a healthcare leader, the playbook is simple:
Establish AI governance frameworks before deployment.
Design for human-AI collaboration with documented workflows keeping humans in the loop for critical decisions.
Start with boring stuff. The organizations succeeding share a common approach: they're starting with use cases that augment human expertise rather than substitute for it.
The organizations winning with AI right now aren't the ones making headlines. They're the ones quietly automating the grunt work while everyone else argues about robot doctors.