Uddin, Nasir (2025) Artificial Intelligence for Intelligent Financial Reconciliation and Discrepancy Detection. American Journal of Economics and Business Management, 8 (5). ISSN 2576-5973

Text AJEBM_Artificial+Intelligence+for+Intelligent+Financial.pdf Download (527kB)

Abstract

Background: The use of artificial intelligence (AI) for financial reconciliation to enhance transaction matching, discrepancy detection, monitoring, and financial control has been increasing. Nonetheless, empirical data on how various AI skills contribute to reconciliation effectiveness is scarce. Methods: In this research, a quantitative approach involving an online survey was used among 168 respondents from the USA who had professional experience in the fields of accounting, finance, banking, auditing, and associated areas. The collected data was analyzed using percentage descriptive and multiple linear regressions to determine the extent of AI usage, operational benefits, effectiveness of discrepancy detection, and AI skills that contribute to reconciliation effectiveness. Results: The results show that there is increasing maturity in AI in reconciliation processes. There were 42.3% adoption in AI Matching and 44.6% adoption in Discrepancy Screening. Early Detection and Matching Accuracy provided great value for 53.6% and 53.0% of the respondents respectively. System Mismatch emerged as the most critical need for improvement at 29.8%. From regression analysis, it was found that the model accounted for 57.8% of the variance in effectiveness of reconciliation. Anomaly Detection, AI Matching, Transaction Monitoring, System Integration, and Explainable AI emerged as the predictors in order. Conclusion: AI can increase the effectiveness of financial reconciliation through better accuracy, discrepancy detection, monitoring, and quick response capabilities. This will require more system integration, consistency in data, explain ability of AI, and proper human supervision.

Item Type:	Article
Uncontrolled Keywords:	Artificial Intelligence, Financial Reconciliation, Discrepancy Detection, Anomaly Detection, Explainable AI
Subjects:	H Social Sciences
Depositing User:	admin eprints
Date Deposited:	01 Sep 2026 16:42
Last Modified:	01 Sep 2026 16:42
URI:	http://eprints.umsida.ac.id/id/eprint/17030

Actions (login required)

View Item