Decision Assurance by Design A Capability Framework for Trustworthy AI Governance in the Enterprise

Artificial intelligence increasingly informs consequential enterprise decisions, yet prevailing governance approaches tend to assess ethical principles, model properties, regulatory conformity or audit activities in isolation. This conceptual paper argues that trustworthy AI governance should focus on the organizational decision supported by AI rather than on the model alone. It develops Decision Assurance by Design, a capability framework that embeds assurance requirements across the AI lifecycle and connects trustworthy AI principles to organizational capabilities, operational controls, assurance evidence and explicit decision authorization. A structured integrative review synthesizes research on responsible AI governance, algorithmic auditing, organizational capability, sociotechnical systems and assurance reasoning. Five mutually reinforcing capabilities are identified: decision purpose, risk and accountability; data and model integrity; human oversight and decision integration; controls, evidence and tra