Explainable Artificial Intelligence (XAI) is crucial in credit scoring to ensure transparency, regulatory compliance, and stakeholder trust. Traditional black-box models, despite their predictive accuracy, often lack interpretability, hindering the explanation of credit decisions. This paper presents a novel framework that integrates decision trees with large language models (LLMs) to enhance explainability in credit scoring applications. The decision tree generates clear decision paths that illustrate how financial attributes affect credit outcomes. These paths are translated into coherent, human-readable explanations using a GPT- based LLM. Our methodology involves data preprocessing, interpretable model training, extraction of decision paths, and LLM integration for explanation generation. The proposed framework achieves a testing accuracy of 71.77%, balancing interpretability with acceptable predictive performance. Qualitative evaluations demonstrate that the generated explanations are logically consistent, relevant, and easily understood by non- technical stakeholders, fulfilling ethical standards and complying with regulations like the General Data Protection Regulation (GDPR)( Rokach, L., & Maimon, O. (2005). “Decision Trees.” In Data Mining and Knowl- edge Discovery Handbook (pp. 165–192). Springer.) ( Doshi-Velez, F., & Kim, B. (2017). “Towards a rigorous science of interpretable machine learning.” arXiv preprint arXiv:17,008,608.). This approach addresses limitations of existing XAI methods by providing human-centered explanations and represents a significant advancement in responsible AI adoption within the financial industry.