Framing the Findings:
This report examines how health systems are advancing along the agentic AI maturity curve, from initial adoption to enterprise-wide optimization. Each section centers on a core dimension for AI-driven transformation in health systems. The insights in this report are grounded in a mixed-method research approach conducted jointly by THMA and Microsoft as part of the AI Transformation Research and AI Collaborative.
Key Takeaways:
Agentic AI in healthcare requires approaching it as digital colleagues, not just tools: Health systems are moving beyond traditional AI applications toward autonomous agents that integrate into workflows, requiring new approaches to human-agent team management and outcome-based alignment.
Governance gaps and ROI uncertainty limit AI scalability: While macro pressures are driving AI investment priorities, practical barriers around governance frameworks constrain ROI demonstration and transformation efforts.
AI is transitioning from pilot programs to foundational infrastructure: There is a shift from experimentation to operationalization; external partners are filling internal capability gaps as AI becomes essential to financial performance, clinical delivery, and workforce strategy over the next 3–5 years.
Three imperatives will determine agentic AI success: governance, data, and workforce: Health systems must simultaneously build continuous governance systems, mature their data infrastructure, and develop workforce capabilities to manage AI agents effectively.
AI maturity and strategy vary significantly across health systems: Perspectives from five diverse AI-first systems reveal different stages of AI adoption, with common barriers to transformation but divergent long-term aspirations.