ADVANCING HEALTHCARE AI GOVERNANCE THROUGH A COMPREHENSIVE MATURITY MODEL

Abstract

The rapid integration of artificial intelligence into healthcare systems has created urgent demand for structured governance approaches that balance innovation with patient safety. While individual hospitals and health systems have begun adopting AI tools for diagnostics, clinical decision support, and administrative automation, governance practices remain inconsistent and fragmented. This paper proposes a comprehensive maturity model for healthcare AI governance designed to help organisations evaluate their current capabilities and progress toward responsible deployment. The model is developed through a mixed-method approach combining a structured review of existing governance frameworks, expert consultation, and empirical assessment across a sample of healthcare organisations. Five maturity levels are defined, ranging from ad hoc practices to fully optimised governance with continuous improvement loops. The framework spans six core dimensions including strategy and leadership, risk management, data stewardship, ethical oversight, workforce readiness, and monitoring. Empirical findings indicate that most surveyed organisations currently sit between levels two and three, with notable weaknesses in post-deployment monitoring and bias auditing. Retrieval-based diagnostic tools and clinical decision support showed the highest governance maturity, while administrative AI applications lagged behind. The paper concludes that achieving robust healthcare AI governance requires coordinated investment in policy, workforce training, and technical monitoring infrastructure, supported by leadership commitment and clear accountability structures.