AI-Powered Predictive Risk Scoring in Continuous Auditing: A Framework for Real-Time Fraud Detection and Compliance in the Banking Sector
Dr. Supriya Arvind Chougule,
Deshbhakta Ratnappa Kumbhar College of Commerce,
Kolhapur.
Abstract
The banking sector is under unprecedented scrutiny from regulators, investors, and customers to ensure transparency, reduce fraud, and maintain compliance. While traditional audit models have been effective in periodic assurance, they are inadequate for detecting emerging risks in real-time. Continuous auditing has emerged as a promising solution; however, most implementations still rely on rule-based monitoring, limiting their ability to identify novel fraud patterns. This paper presents a conceptual framework for integrating Artificial Intelligence (AI)-driven predictive risk scoring into continuous auditing systems, with a focus on the banking sector. The proposed model leverages machine learning, anomaly detection, and natural language processing to assign dynamic risk scores, detect suspicious activities, and ensure compliance with regulatory guidelines such as Reserve Bank of India (RBI) directives, Basel III, and the Sarbanes–Oxley Act (SOX). The framework also addresses explainability, bias mitigation, and data privacy concerns. By moving from reactive to proactive risk management, AI-powered predictive risk scoring has the potential to transform audit practices in the financial services industry.
Keywords: Artificial Intelligence, Predictive Risk Scoring, Continuous Auditing, Fraud Detection, Banking Sector, Regulatory Compliance, Machine Learning, Anomaly Detection
DOI link – https://doi.org/10.69758/GIMRJ/2509I9VXIIIP0068
Download