AI-Driven Credit Risk Assessment and Financial Inclusion in Emerging Markets: Evidence from Pakistani Firms

Abstract

Financial exclusion is an enigma in emerging economies, where existing credit risk assessment systems centre on formal credit history and collateralisation, systematically locking out small firms, informal businesses, and first-time borrowers. The latest trends in artificial intelligence (AI) offer new opportunities to revolutionize credit risk assessment through alternative data and innovative predictive analytics. The paper focuses on the immediate impact of AI-based credit score models on financial inclusion in the context of Pakistani companies that engage in lending and credit rating. The research article draws on information asymmetry and financial intermediation theories and argues that AI-guided credit scoring increases financial inclusion by enabling lenders to assess creditworthiness beyond traditional information constraints. Using a quantitative, cross-sectional research design, primary data were collected from 265 workers in Pakistani banks, microfinance institutions, FinTech companies, and SME-focused lending institutions. Partial Least Squares Structural Equation Modeling (PLS-SEM) was used to empirically test the proposed model. The findings indicate that AI-based credit scoring has a significant positive direct impact on financial inclusion. This paper adds to the literature on FinTech and financial inclusion by presenting firm-level empirical data from a poorly studied emerging market: Pakistan. The results also have significant implications for financial institutions and policymakers by demonstrating the strategic value of AI-powered credit assessment models for expanding access to finance and, at the same time, for effective risk management.