The Impact of Artificial Intelligence Application on Fraud Detection in Financial Statements: The Mediating Role of Accounting Information System Quality

The purpose of the current study was to explore the effect of artificial intelligence (AI) through its components of machine learning and data mining on the detection of fraud in financial statements with consideration of the mediating role of AIS quality in this context. To attain the objectives of this study, a descriptive-analytical methodology was used. A structured questionnaire was employed as the primary data collection instrument and distributed to a sample of employees working in financial institutions. The collected data were analyzed using the Statistical Package for the Social Sciences (SPSS) through a range of appropriate statistical techniques. The findings revealed that the application of artificial intelligence had a statistically significant effect on fraud detection in financial statements. Specifically, machine learning techniques explained 70.7% of the variance in fraud detection, whereas data mining tools accounted for 83.2% of the observed variance. The results also demonstrated a statistically significant effect of AI applications on the quality of accounting information systems, with machine learning techniques explaining 39.5% of the variance and data mining tools explaining 46.2% of the variance in AIS quality. Furthermore, the quality of accounting information systems was found to have a statistically significant effect on financial statement fraud detection. However, the results of the mediation analysis did not provide sufficient statistical evidence to support a significant mediating role of accounting information system quality in the relationships between machine learning techniques and financial statement fraud detection or between data mining tools and financial statement fraud detection, as the 95% Bootstrap confidence intervals for the indirect effects included zero. Based on these findings, the study recommends strengthening the adoption and direct application of artificial intelligence technologies within private banks, particularly machine learning techniques and data mining tools, given their significant contribution to enhancing financial statement fraud detection. It also recommends continuously improving the quality of accounting information systems by enhancing data accuracy, reliability, timeliness, and structure to support effective financial monitoring and control. However, since the mediating role of accounting information system quality was not statistically supported, banks should not rely on AIS quality as the primary mechanism through which artificial intelligence improves fraud detection. In its place, more focus needs to be put on the actual incorporation of artificial intelligence techniques in fraud detection, auditing, and financial controls.