AI agents are rapidly spreading in health care, but validation is lacking

March 11, 2026 - 06:54

The annual HIMSS conference has become a showcase for a new wave of artificial intelligence, with AI agents promising to automate complex healthcare tasks appearing in nearly every booth and presentation. These digital assistants are designed to perform duties ranging from clinical documentation and prior authorization to offering diagnostic support and managing patient communication.

However, amidst the enthusiastic demonstrations, a critical question is being raised by clinicians and health IT leaders: where is the rigorous validation? Experts are expressing concern that the breakneck speed of development and deployment is outpacing the essential process of proving these agents are safe, effective, and equitable. Many tools are entering the market with limited peer-reviewed research or real-world clinical trials to substantiate their claims.

The lack of standardized validation frameworks means healthcare systems are often left to assess these complex tools on their own, a risky and resource-intensive proposition. This validation gap creates potential pitfalls, including the perpetuation of biases, patient safety risks, and integration failures that could disrupt care workflows. The industry consensus is clear: for AI agents to move from promising prototypes to trusted partners in care, developers must prioritize transparent, third-party validation to build the essential foundation of trust with the medical community.