AI-BASED CREDIT RISK ASSESSMENT IN THE DEVELOPMENT OF DIGITAL BANKING SERVICES

Annotatsion: This article develops a methodology for assessing commercial banks’ credit risk using artificial intelligence (AI) technologies amid the expansion of digital banking services. The study draws on publicly available statistics of the Central Bank of the Republic of Uzbekistan for 2024–2025, international prudential recommendations, and a computational experiment based on a synthetic micro-dataset of 10,000 loan applications. Logistic regression, decision tree, Random Forest, and XGBoost models were compared on a stratified test sample using ROC-AUC, F1, KS, Brier score, and calibration criteria.