Explainable AI in Banking Compliance: Leveraging Large Language Models for AML and KYC Decision Support

In this study, we propose an explainable artificial intelligence framework for banking compliance that integrates traditional machine learning models with large language models to support Anti-Money Laundering and Know Your Customer decision-making. The proposed approach emphasizes regulatory transparency, auditability, and human-centered interpretability while maintaining strong predictive performance. Using an open-source financial dataset from the UCI Machine Learning Repository, we demonstrate how explainable modeling and natural language reasoning can enhance compliance decision support systems. This research aligns with U.S. national interests by improving financial system integrity, reducing compliance costs, and strengthening risk monitoring capabilities in regulated institutions.