Artificial Intelligence in Forensic Accounting: A Bibliometric Analysis of Global Research Trends and Future Directions

Abstract

Artificial Intelligence (AI) has significantly transformed forensic accounting practices, particularly in supporting fraud detection and financial investigations. As the number of publications in this field continues to grow, a comprehensive mapping of the literature is needed to understand the global evolution of research. This study aims to examine the development of research on AI in forensic accounting using a bibliometric approach. Bibliographic data were retrieved from the Scopus database following the PRISMA framework. After the screening process, 399 articles published between January 2015 and July 2026 were analyzed using Biblioshiny (Bibliometrix) in RStudio. The findings reveal a consistent increase in research output, with an annual growth rate of 42.84%, particularly since 2022. China emerged as the leading contributor in terms of both publications and citations, while Knowledge-Based Systems and Decision Support Systems were identified as the most productive journals in this research area. Keyword analysis indicates that machine learning, fraud detection, artificial intelligence, deep learning, and financial fraud remain the dominant research themes, reflecting an increasing emphasis on advanced AI technologies. Furthermore, Lotka's Law suggests that this research field is still in its growth stage, whereas Bradford's Law indicates that publications are concentrated within a relatively small number of core journals. This study provides a comprehensive overview of the evolution of AI research in forensic accounting and identifies several promising directions for future research, including Explainable AI, Responsible AI, Large Language Models, and AI governance.