Generative artificial intelligence in model risk management: emerging opportunities, supervisory challenges and validation frameworks

The rapid evolution of generative artificial intelligence (GenAI) has catalyzed a paradigm shift in model development, interpretability and governance across the financial services sector. Large language models and multimodal architectures are increasingly embedded in risk management, compliance and operations, creating both transformative opportunities and unprecedented supervisory challenges. This paper examines the intersection of GenAI and model risk management in relation to emerging regulatory expectations, model validation frameworks and practical applications. Drawing on recent supervisory developments and empirical experience in model risk governance, it proposes a structured approach to validating GenAI systems in line with the principles of US Federal Reserve Supervisory Letter SR 11-7, Prudential Regulation Authority Supervisory statement Supervisory Statement SS 1/23 and Basel Committee on Banking Supervision AI risk management guidance. The paper also presents a real-world case study on validating an alert-reduction model for sanctions screening, based on a large language model, which illustrates how conceptual soundness, performance testing and interpretability can be achieved in an environment where regulatory expectations, explainability and innovation intersect.