Rural Hospitals Face AI Investment Dilemma Amid Medicaid Cuts

Rural health systems across the United States are confronting a stark contradiction: federal officials are promoting artificial intelligence as the solution to financial pressures, while hospital leaders on the ground say the technology is too expensive and its cost-saving potential remains unproven.

The disconnect comes as rural providers face nearly $1 trillion in Medicaid cuts over the next decade. Centers for Medicare and Medicaid Services administrator Mehmet Oz has championed AI-based solutions, including virtual avatars for mental health care, as the path forward. The administration argues that a $50 billion rural health transformation fund accompanying the cuts will enable AI adoption that could fundamentally remake care delivery and reduce costs.

The cost reality

Health system executives serving rural communities paint a different picture. Lori Dwyer, president and CEO of Penobscot Community Health Care in Maine, said her organization has already implemented AI-powered ambient scribes that listen to patient conversations and help with electronic health record documentation.

The technology has reduced administrative burden and improved clinician satisfaction, but "that doesn't reduce costs," Dwyer told STAT, which first reported these details. Other AI applications her system uses for patient communications and remote monitoring have created only marginal economic efficiencies—insufficient to offset the looming budget cuts.

Across dozens of interviews with rural providers, hospital leaders, and health AI experts, a consistent theme emerged: AI tools require significant upfront investment, and their return on investment has not been demonstrated at scale. Many system leaders expressed concern that large, well-resourced health systems will capture most benefits from AI adoption, potentially widening existing health inequities rather than closing them.

Infrastructure barriers

Rural areas face additional obstacles beyond cost. Limited broadband access in states like Maine can make internet-dependent AI systems unreliable. Many rural facilities still rely on faxed or handwritten forms that cannot easily integrate with digital systems. Even digitized records often require extensive reorganization before AI tools can process them effectively.

"If you're employed by a big health system, you're much more likely to have that opportunity," said James Jarvis, president of the Maine Medical Association. "If you're at a private practice, you're much less likely to have that advantage."

James McHugh, a managing director at Berkeley Research Group who helps clients implement health technology, noted that many AI implementations have struggled and that labor cost reductions "usually doesn't happen" despite vendor promises.

Acceleration despite doubts

Some rural health leaders are nevertheless accelerating AI adoption, viewing it as essential for survival. Trampas Hutches, Mountain Region president at MaineHealth, said the system is working to make AI a "care team member" while using it for documentation and administrative tasks. More than 70% of hospitals now use some form of predictive AI, according to survey data.

Yet even advocates acknowledge challenges. "There's a lot of garbage out there," McHugh said of available AI products. Hutches compared the current market to the dot-com bubble, noting the difficulty of identifying worthwhile investments.

Why it matters

The gap between federal AI promotion and rural implementation reality has significant implications for health equity. If only well-funded systems can afford effective AI adoption, technology could accelerate the consolidation of care in urban centers and widen disparities in rural access. The outcome will test whether emerging technology can genuinely democratize quality care or simply create new divides between resource-rich and resource-poor providers.

STAT reporter Daniel Payne traveled to Maine and Louisiana and interviewed health AI experts and vendors for this reporting.