Hackensack Meridian Health Completes First Joint Commission AI Certification in U.S.

Hackensack Meridian Health has completed the first Joint Commission AI certification process in the United States, establishing a national benchmark for how health systems should govern the clinical AI tools embedded across their operations. The certification signals a structural shift in healthcare AI accountability — from voluntary guidelines and self-reported compliance to structured third-party audits with accreditation consequences.

The Joint Commission is the dominant hospital accreditation body in the United States. Its standards carry real weight: hospitals that lose accreditation lose Medicare and Medicaid reimbursement eligibility. According to HISTalk's September 23 healthcare AI briefing, the certification process reviewed Hackensack Meridian's full inventory of clinical AI tools, risk assessments, ongoing monitoring protocols, and patient data protections — a comprehensive audit scope that goes well beyond what most hospitals have done voluntarily.

What the Certification Covered

Hackensack Meridian Health is one of New Jersey's largest health systems, operating 17 hospitals. Its AI footprint spans clinical decision support — tools that flag sepsis risk, predict deterioration, suggest diagnoses — as well as administrative AI covering prior authorization, documentation, and scheduling.

The certification process required the health system to:

• Maintain a complete inventory of every clinical AI tool in production, including vendor-supplied algorithms embedded in imaging, EHR, and monitoring systems
• Conduct and document formal risk assessments for each tool, including bias testing, performance monitoring, and failure mode analysis
• Demonstrate ongoing monitoring protocols — not just pre-deployment validation, but continuous post-deployment tracking of model performance against real patient outcomes
• Show patient data protection practices that meet both HIPAA requirements and the specific data governance expectations the Joint Commission has codified for AI systems

The full scope of the Joint Commission's AI standards has been published as part of its evolving accreditation framework, representing the formal operationalization of principles that hospitals have been trying to self-certify against for years.

Why This Matters Beyond One Health System

Healthcare AI has faced a persistent accountability gap. The FDA clears AI-based medical devices through its software-as-a-medical-device (SaMD) pathway, but the vast majority of clinical AI tools — particularly those providing recommendations rather than diagnoses — fall outside FDA jurisdiction. Hospitals have been buying and deploying these tools with minimal external oversight.

Voluntary frameworks from organizations like the American Hospital Association, the American Medical Association, and ECRI have provided guidance, but guidance without accreditation consequences has limited effect on procurement and deployment behavior. The Joint Commission's entry changes the incentive structure.

What Hackensack Meridian proved possible: a major health system can submit its AI governance practices to external audit and pass. That removes the "this isn't ready yet" argument that has slowed adoption of formal certification in healthcare.

The Broader Accreditation Signal

The Joint Commission does not typically move fast. Its decision to develop and execute an AI certification process reflects how far clinical AI has penetrated health system operations — and how much liability exposure health systems now face from AI errors. Several hospitals have already faced malpractice litigation involving AI-flagged or AI-missed diagnoses. Formal certification creates both a defense ("we followed the accredited standard") and a floor below which negligence becomes harder to dispute.

Health system executives watching this certification now face a practical question: do you want to be audited before an adverse event — or after one?

What to Watch

Watch whether the Joint Commission issues updated standards requiring all accredited hospitals to conduct AI inventories as a baseline condition. That move — possible in 2027 — would create a compliance mandate for every U.S. hospital that holds Joint Commission accreditation, an estimated 22,000 facilities. Also watch how large health system AI vendors respond: if their tools need to be certified-friendly to clear a hospital's accreditation review, vendor documentation and monitoring capabilities become procurement requirements, not optional features.