The recent controversy over artificial intelligence tools in rural health care began with a single word: “avatar.” That single term sidetracked the conversation, shifting attention away from the far more urgent issue of how to sustain primary care in communities that are steadily losing it.
Dr. Mehmet Oz, administrator of the Centers for Medicare and Medicaid Services, used that reference in remarks about the severe shortage of clinicians in underserved communities, suggesting that AI-enabled tools could help extend care in rural America. He emphasized that physicians would remain involved and that technology should support care teams, not replace them. Yet much of the reaction focused on the label rather than the underlying problem.
CMS later clarified the premise: AI should be explored as a tool to extend the reach of licensed clinicians, operating under clinical oversight with clear safeguards for safety, privacy and accountability. That is the right frame.
Lost in the semantic swirl is a measurable and worsening primary care crisis.
Roughly 60 million Americans live in rural communities. Many reside in areas designated as primary care Health Professional Shortage Areas. Since 2005, more than 190 rural hospitals have closed. Rural residents face higher rates of heart disease, stroke, chronic lung disease and diabetes than their urban counterparts. The Association of American Medical Colleges projects significant physician shortages in the coming decade, with primary care among the most strained specialties.
Primary care clinicians in these communities are stretched thin. In addition to seeing patients, they manage documentation, prior authorizations, inbox messages, preventive outreach and chronic disease follow-up. Administrative burden consumes hours each day. When capacity shrinks, access narrows. When access narrows, continuity erodes. When continuity erodes, outcomes suffer.
Artificial intelligence, if implemented responsibly, can help address that imbalance.
The most appropriate applications are structured and clinician-supervised. AI can assist with standardized intake and scheduling. It can automate reminders and outreach to close preventive care gaps. It can facilitate between-visit check-ins that surface straightforward changes in stable chronic conditions. It can generate draft documentation for clinician review, reducing after-hours charting and clinician burnout.
These uses extend capacity. They do not replace clinical judgment.
What determines safety is not the label attached to a patient-facing interface but the governance behind it. Any AI-enabled workflow must include identity verification, validated clinical content and clearly defined triage protocols. High-risk symptoms or ambiguous presentations must automatically escalate to a licensed clinician. AI should not independently diagnose. It should not make emergency decisions. It should not present itself as a licensed provider.
Accountability remains with clinicians and the organizations responsible for care delivery. Outputs must be auditable and continuously monitored. Patients deserve transparency about when AI is involved, what it does and how to reach a human being quickly. Data use must adhere to minimum-necessary standards with strong privacy protections. Equity must be central, accounting for broadband gaps, phone-first preferences, language access and varying levels of digital literacy in rural communities.
This is not about adopting technology for its own sake. It is about protecting the viability of primary care as a profession and as a public good. If we do not reduce administrative friction and expand capacity, more rural clinicians will retire early, scale back their panels, or leave underserved areas altogether. No software can replace a trusted relationship between a patient and a doctor, but thoughtful tools can help preserve that relationship by giving clinicians back time.
Importantly, policy now has resources attached to it. CMS has begun distributing $50 billion in rural health transformation funds authorized in the so-called Big Beautiful Bill. Those dollars must do more than stabilize struggling facilities. They should catalyze structural change in how rural primary care is delivered, financed and supported. Investments in workforce, workflow redesign and clinician-supervised technology must be tied to measurable improvements in access, continuity and outcomes. Funding without transformation will not solve the crisis.
Oz, CMS, doctors and patients can agree on one thing: We do not have enough primary care clinicians. The solution is not substitution. It is an extension under supervision.
The rural health crisis is real. The label debate isn’t. It is time to move beyond semantics and focus on governed innovation that strengthens primary care teams, protects patients and creates more human time in care where it is needed most.
About Dr. David Carmouche
Dr. David Carmouche is executive vice president and chief clinical transformation officer at Lumeris, where he leads clinical strategy and AI-enabled primary care transformation, following prior senior leadership roles overseeing care delivery at Walmart Health, Ochsner Health and Blue Cross Blue Shield of Louisiana.
