AI-governed audit assurance: A conceptual framework integrating predictive intelligence, governance controls, and continuous assurance

The adoption of Artificial Intelligence (AI) in audit assurance enables the identification, automation and analysis of auditor risks in real-time. However, new challenges and risks have also emerged, for example, on issues of transparency, accountability, auditability, and governance. This study proposes an AI-governed audit assurance framework that integrates predictive modeling, audit analytics, and AI governance. The framework is a three-layered system with predictive intelligence, governance control modules, and an audit assurance layer, addressing the gap in conceptual integration. This paper describes an AI audit governance layer supporting model validation, explainability, bias detection, disclosure governance and compliance governance. The paper describes how the governance layer can be integrated into the lifecycle of AI and deep learning models. A framework for measuring assurance is presented along with a roadmap for implementation. The paper contributes to the field by rethinking predictive models as governed assurance objects. It contributes by integrating AI and deep learning models into a broader governance framework. The paper identifies key elements of an AI audit governance framework. It highlights audit assurance as a critical governance mechanism enabling trust in organizations’ predictive systems and models.