Rethinking the Role of Internal Audit in the Era of Artificial Intelligence: A Critical Review of Empirical Studies and Future Research Directions

Abstract

The rapid advancement of Artificial Intelligence (AI) and Machine Learning (ML) is progressively transforming professional auditing practices, particularly the internal audit function, which has traditionally relied on manual, sample-based, and retrospective procedures. This article presents a critical review of the empirical literature on the integration of Artificial Intelligence into internal auditing, drawing on a corpus of international studies published primarily between 2019 and 2026 in journals indexed in Scopus and Web of Science. Because studies focused exclusively on the internal audit function remain relatively scarce, part of this corpus originates from research on external audit or on auditing and accounting in general; such contributions are mobilised with due methodological caution and are systematically distinguished from those specifically addressing internal audit. The study pursues two main objectives: first, to provide a structured overview of the contributions, limitations, and determinants of AI adoption within the internal audit function; and second, to identify future research directions that may guide subsequent investigations, particularly in underexplored contexts such as developing countries. The review of recent systematic literature studies, selected according to explicit criteria of methodological rigour, recency and disciplinary complementarity, highlights the principal theoretical foundations underpinning this field, including the Technology-Organization-Environment (TOE) framework, the Resource-Based View (RBV), the Diffusion of Innovation (DOI) theory, and the Technology Acceptance Model (TAM). The findings reveal a broad consensus regarding the positive impact of AI on audit efficiency, accuracy, and the detection of anomalies and fraud, although this consensus should be read cautiously given the methodological heterogeneity of the underlying studies. However, they also indicate that this technological transformation raises significant ethical, regulatory, technical, and societal challenges that continue to hinder the widespread adoption of AI by internal auditors. These findings underscore the need for further research into the organizational and human conditions required for the successful integration of AI technologies into internal auditing, particularly in the Moroccan context, where research on this topic remains at an early stage.