AI-Driven ESG Credit Scoring for SMSs in Islamic Finance: A Conceptual Framework

Abstract

Small and medium-sized enterprises (SMEs) frequently face financing constraints because conventional credit assessment depends heavily on audited financial statements, established credit histories and collateral. These requirements may disadvantage viable but information-poor SMEs. This conceptual paper develops a literature-derived framework explaining how alternative SME data may be translated into environmental, social and governance (ESG)-related signals and incorporated into artificial intelligence (AI)-assisted credit assessment within Islamic finance. The paper adopts a structured conceptual approach that integrates literature on SME information asymmetry, alternative-data credit assessment, proportionate ESG evaluation, explainable AI and Maqasid al-Shariah-based governance. The framework distinguishes raw operational data from credible business signals and ESG-related indicators. Signaling Theory explains how verifiable, relevant and context-sensitive operational records may reduce information asymmetry, while AI is positioned as a decision-support mechanism that integrates validated alternative signals with conventional financial information. Maqasid al-Shariah operates as an evaluative governance layer guiding data relevance, consent, fairness, explainability, human oversight and institutional accountability. The paper contributes by specifying the conceptual relationships connecting alternative data, ESG-signal formation, AI-assisted analysis and Islamic ethical governance. However, the framework has not undergone expert assessment or empirical validation. It therefore cannot claim improved predictive accuracy, fairness, financial inclusion or Shariah compliance. Instead, it provides a structured set of propositions for future operationalisation, testing and controlled institutional assessment.