A review of credit risk assessment in banking - from deterministic models to probabilistic AI

This paper examines the literature on applications of artificial intelligence (AI) and machine learning (ML) in assessing credit risk in banks, specifically focusing on probabilistic AI. We uncover unresolved issues and prevalent technical challenges by identifying the predominant AI and ML models employed. Furthermore, we explore potential approaches to address these challenges. The review reveals that research is increasingly exploring AI and ML models for credit risk modeling, often providing more accurate risk predictions than traditional models. Apparently, there is a predominant reliance on traditional deterministic models, highlighting an underutilization of probabilistic models that effectively handle inherent uncertainties in credit risk estimates. Furthermore, we observe that challenges such as overfitting, model interpretability, and regulatory compliance remain prevalent. This suggests significant potential for improving accuracy, reliability, and trustworthiness by increasing applications of probabilistic and generative AI, in combination with explainable AI (XAI). • Through a systematic literature review, we find that probabilistic AI models largely remain underutilized in the scientific literature. • We argue that probabilistic AI models are well-suited to address the current lack of transparent and interpretable models that comply with regulatory requirements. • We recommend that researchers conduct in-depth explorations of probabilistic models in order to more accurately capture both epistemic and aleatoric uncertainty. • This should be accompanied by further research on XAI and generative AI as means to enhance the transparency and interpretability of models.