Artificial intelligence and credit risk assessment in Nigeria’s digital lending market: an econometric analysis

Abstract

The quick growth of Nigeria’s digital lending market has brought about the efficient credit risk assessment mechanisms. The Traditional credit scoring models faced many challenges such as the limited credit histories and high information asymmetry. This study examines the impact of Artificial Intelligence (AI)–driven credit scoring on loan default rates in Nigeria’s digital lending ecosystem using simulated panel data reflecting fintech lending operations between 2017 and 2025.they used the Information Asymmetry Theory and Financial Intermediation Theory in anchoring the study and employs a logistic regression and panel Ordinary Least Squares (OLS) techniques to evaluate the relationship between AI adoption and credit performance. The findings shows that the AI-based credit scoring significantly reduces default probability by approximately 18–24% compared to traditional models. Alternative data utilization shows positive and statistically significant effects on predictive accuracy. However, regulatory compliance intensity moderates these effects. The study concludes that AI enhances predictive efficiency and financial inclusion but requires robust governance and regulatory oversight.