Explainable Artificial Intelligence in ESG analytics for sustainable finance: a systematic review

Explainable Artificial Intelligence (XAI) is increasingly transforming ESG analytics in sustainable finance by improving the transparency, interpretability, and trustworthiness of AI-driven decision-making. While machine learning has significantly enhanced ESG prediction, disclosure assessment, and sustainable investment analysis, limited model explainability remains a critical barrier to stakeholder trust, regulatory compliance, and practical adoption. This study systematically reviews the literature on Explainable Artificial Intelligence in ESG analytics using the Theory, Context, Characteristics, and Methodology (TCCM) framework, supported by the Scientific Procedures and Rationales for Systematic Literature Reviews (SPAR-4-SLR) protocol. Drawing on peer-reviewed studies indexed in the Scopus and Web of Science databases, the study combines bibliometric and content analyses to examine publication trends, theoretical foundations, contextual settings, methodological approaches, and emerging research themes. The findings indicate a growing transition from conventional machine learning toward XAI in ESG analytics, with increasing adoption of techniques such as SHAP, LIME, and other interpretable learning approaches to enhance transparency, accountability, and stakeholder trust in ESG-related financial decision-making. The review also reveals the methodological dominance of quantitative research, with comparatively limited qualitative and mixed-method studies. By synthesizing the intersection of XAI, ESG analytics, and sustainable finance, this study develops a future research agenda emphasizing explainability, responsible AI governance, model transparency, and sustainable financial innovation. The study advances knowledge on how XAI strengthens ESG credibility, supports investment decisions, and promotes transparent, sustainable financial systems.