V3I7P7

AI-Driven Sustainable Banking: Examining the Role of Artificial Intelligence in Green Credit Assessment

Satyam Srivastava¹*, Lakshmi Narashiman², Kanchan G. Rajput³

Abstract

Despite the potential of artificial intelligence to revolutionize green credit assessment, little information exists on its application in sustainable banking. The present research aims to explore the ways in which AI approaches can be used to factor in environmental risks in loans. The first identified research gap lies in a scarcity of empirical analyses concerning the inclusion of climate risk measurement through AI algorithms alongside more traditional credit score inputs. Utilizing literature across academia and industry (from 2015 to 2026), this section synthesizes empirical findings on climate finance, ESG scores, and AI-driven risk assessment. Empirical findings include evidence that the use of AI allows banks to punish greenwashing firms and that a higher proportion of green loans may lower NPLs. In doing so, the progress made in the field is juxtaposed against traditional approaches and described in terms of data collection (such as company reports and satellite data) and feature extraction. A framework for AI-driven lending is outlined in terms of how capabilities (including big data and machine learning) may enable climate risk evaluation and better loaning decisions. Finally, we provide an overview of an empirical methodology for utilizing AI in the lending process. This includes data gathering, from structured financial and ESG metrics to unstructured sources such as satellite images and textual analysis, alongside model training and dual metric evaluation (credit risk and environmental performance). Explainability methods are incorporated throughout the framework.

Keywords:

Artificial Intelligence, Green Credit Assessment, Sustainable Banking, Climate Risk, ESG Scores, Machine Learning, Green Finance, Explainable AI