A Study on Predicting ESG Rating Downgrades of Korean Companies Post-COVID-19 using Explainable Artificial Intelligence

The importance of environmental, social, and governance (ESG) performance and related risk management has grown significantly in the aftermath of the COVID-19 pandemic. This study aims to predict ESG rating downgrades among publicly listed firms in Korea during the post-pandemic period (2020-2024) by comparing the performance of various machine learning models. Furthermore, explainable artificial intelligence (XAI) techniques are employed to interpret the prediction outcomes of top-performing models in greater depth. By integrating XAI methods, this study addresses the “black box” nature of AI-driven ESG prediction models, enhancing both interpretability and credibility of the results. The findings are expected to offer empirical insights for corporate managers, investors, and policymakers to proactively identify and manage ESG-related risks and make more informed, data-driven decisions.