AN INVESTIGATION OF BASEL III COMPLIANCE AND ARTIFICIAL INTELLIGENCE–DRIVEN CREDIT RISK MANAGEMENT IN THE U.S. BANKING SECTOR

Abstract

This study investigates the role of Basel III compliance and Artificial Intelligence (AI)–driven credit risk management in strengthening financial stability within the United States banking sector. In the increasingly complex digital financial environment, U.S. banks face significant challenges in accurately assessing credit risk due to large-scale data generation, evolving borrower behavior, and market volatility. The research problem focuses on the limited integration between Basel III regulatory frameworks and AI-based predictive systems in improving credit risk accuracy, early default detection, and risk mitigation efficiency across U.S. financial institutions. The study adopts a quantitative research methodology using secondary data collected from annual reports of major U.S. commercial banks, Federal Reserve regulatory filings, and financial databases covering the period 2019–2025. The dataset includes key credit risk indicators such as non-performing loan ratios, capital adequacy ratios, probability of default (PD), loss given default (LGD), and AI-based credit scoring outputs. Statistical techniques, including regression analysis and comparative evaluation, are applied to assess the relationship between Basel III compliance levels and AI-enhanced credit risk management performance. The findings indicate that U.S. banks integrating AI-driven analytics with Basel III frameworks demonstrate improved credit risk prediction accuracy, lower default rates, and stronger capital adequacy positions. AI systems also enhance early warning mechanisms and improve lending decision efficiency. Measurable outcomes include reduced non-performing loan ratios, improved risk-adjusted returns, enhanced liquidity stability, and increased predictive accuracy in credit scoring models.