The Use of Artificial Intelligence in Accounting Auditing

This article examines the growing role of artificial intelligence (AI) in accounting auditing, the opportunities it presents, methodologies, and challenges encountered from an academic perspective. With the proliferation of Industry 4.0 and big data, sampling-based approaches used in traditional auditing are becoming insufficient, creating a need to analyze the entire data universe. In this context, AI provides auditors with unprecedented capabilities to detect fraud, errors, and irregularities within large data sets, forming the basis of the “Audit 4.0” paradigm. The article addresses the three main areas of AI application in auditing—process automation, risk assessment, and decision support systems—within a theoretical framework. Anomaly detection-focused unsupervised learning methodologies, particularly clustering analysis and deep learning-based autoencoder architectures, are examined at a technical level. However, the practical application of AI also brings significant challenges, such as data imbalance, the interpretability of complex models (the “black box” problem), and potential biases in algorithms. As solutions to these challenges, Explainable Artificial Intelligence (XAI) techniques and the concept of “Trustworthy Artificial Intelligence” are discussed. In conclusion, it is emphasized that AI is not a technology that eliminates the auditor, but rather an “augmented intelligence” tool that enhances their capabilities. It is noted that the auditor's role is transforming from manual data verification to that of an expert who interprets AI models and oversees algorithmic systems. This transformation will shape the future of the auditing profession.