Companies above a certain size that operate globally are showing increasing commitment to ESG (environmental, social, and governance) activities. The main goal of this study is to design a model that can predict future corporate value based on ESG score data. To this end, this study compares the predictions of the basic future corporate value prediction model on which previous studies have been based and those of the future corporate value prediction model proposed herein that includes ESG ratings. For a more rigorous analysis that obtains more comprehensive results, the current study presents results using five machine learning methods: CatBoost, Extra Trees, LGBM, Random Forest, and Gradient Boost. These results indicate that models that encompass ESG data consistently outperform models that do not encompass ESG data in terms of predicting future corporate value. This paper is characterized by its use of an interdisciplinary research methodology that uniquely introduces machine learning techniques, which are rarely used for empirical analysis in the financial and accounting fields. This innovative and future-oriented research method is expected to inspire subsequent scholars in these domains and others in which machine learning techniques are not typically used.