AI-DRIVEN CREDIT SCORING AND FINANCIAL INCLUSION: BENEFITS AND RISKS FOR UNDERBANKED POPULATIONS

Traditional credit scoring models often exclude individuals without formal banking histories, creating a persistent barrier to financial inclusion for underbanked populations. This paper examines the paradigm shift toward Artificial Intelligence (AI)-driven credit scoring, which utilizes alternative data—such as mobile phone usage, utility payments, and digital footprints—to assess creditworthiness. The study explores the dual nature of this technological integration: its profound benefits in democratization of credit, reduced operational costs, and expanded financial access, contrasted against critical risks including algorithmic bias, data privacy violations, and the lack of model transparency (the "black box" problem). Through a comprehensive conceptual analysis, the paper highlights the macroeconomic implications of AI in credit markets and proposes a balanced regulatory framework to protect vulnerable consumers while fostering financial innovation.