Healthcare organizations are showing growing interest in agentic AI, but most remain in the early stages of adoption, according to new research from Microsoft and the Health Management Academy.
The findings, published in the January 2026 issue of The New England Journal of Medicine, are based on surveys and in-depth interviews with senior healthcare executives across U.S. provider organizations. The research examines how health systems are progressing along what the authors describe as an “agentic AI maturity curve,” ranging from early experimentation to enterprise-level optimization.
Here are five key findings from the research:
- Forty-three percent of respondents reported piloting or testing agentic AI, but only 3% said they have deployed AI agents in live workflows, according to a Feb. 12 press release from Microsoft.
- At the same time, one-third of respondents said they have no plans to explore agentic AI within the next one to two years.
- Sixty percent of respondents agreed or strongly agreed that agentic AI will meaningfully improve or disrupt the provider-patient experience. Fifty-seven percent expressed similar optimism about productivity gains.
- More than three-quarters of respondents, or 77%, expect AI agents to improve backend productivity, while 60% believe they will fundamentally reshape the patient-provider experience.
- However, leaders cited workforce challenges as adoption expands. Sixty percent identified reskilling and upskilling as a top concern as ecosystems of AI models and agents grow.
At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.
