Pengaruh Machine Learning dalam Mitigasi Risiko Kredit, Deteksi Fraud dan Mitigasi Kerugian Finansial Perbankan Digital

This study aims to analyze the effect of Machine Learning (ML) adoption on the effectiveness of credit risk mitigation and fraud detection in the digital banking sector, focusing on Bank Jago for the 2024-2025 period. The increasing digital transaction phenomenon demands a more precise security and risk management system than conventional methods. Using a quantitative approach with simple linear regression analysis through SPSS, this study examines the effect of ML Adoption (X) on three dependent variables: Non-Performing Loan Ratio (Y1), Fraud Detected (Y2), and Prevented Losses (Y3). The results show that ML Adoption has a significant negative effect on the NPL Ratio. Furthermore, ML has a strong positive significant effect on fraud detection and the value of prevented losses. These findings support Stewardship Theory, where management uses intelligent technology as an instrument to protect customer interests and maintain the company's financial stability. This study concludes that ML integration is a key determinant in maintaining the health of assets and the security of the digital banking ecosystem in the future.