LITERATURE REVIEW: LEVERAGING ARTIFICIAL INTELLIGENCE IN AUDITING FOR DETECTING FRAUD

Abstract

This article examines the role of Artificial Intelligence in auditing with a specific focus on fraud detection, drawing on findings from a systematic literature review and bibliometric analysis. The study highlights how the rapid growth of digital data has made traditional sampling methods less effective, increasing the need for AI-based tools capable of analyzing full populations of transactions. The literature shows that machine learning models such as CART, neural networks, and ensemble techniques significantly improve anomaly detection accuracy while reducing audit processing time. Using the PRISMA framework, the analysis identifies publication trends, dominant authors, key institutions, and frequently occurring keywords related to AI and fraud detection. The results reveal that AI enhances audit quality by identifying patterns that are difficult for manual procedures to capture, but its effectiveness depends on cybersecurity readiness, auditor digital competence, and overall organizational support. Although AI improves efficiency, human judgment remains essential for interpreting results and assessing qualitative factors that algorithms cannot evaluate. The study concludes that AI will continue to play an important role in fraud detection, provided that organizations strengthen their digital infrastructure, ensure proper training, and integrate technology with sound governance practices.