Enhancing transparency and reliability in generative AI through case-based reasoning and explainable AI

Integrating transparency and reliability into generative artificial intelligence remains a critical challenge in modern digital systems due to the opaque and probabilistic nature of deep learning models. This study proposes a hybrid framework that integrates Generative AI, Case-Based Reasoning (CBR), and Explainable AI (XAI) to enhance factual reliability, interpretability, and user trust. The framework is implemented using a transformer-based generative model combined with a semantic-contextual case retrieval mechanism, adaptive case refinement, and dual-layer explainability incorporating both feature attribution and case-based justification. Unlike existing state-of-the-art approaches that focus independently on generation quality or explainability, the proposed framework jointly optimizes reasoning transparency, contextual grounding, and adaptive learning within a unified architecture. Experimental evaluation was conducted on benchmark datasets including MMLU, HotpotQA, e-SNLI, and a custom case-based dataset, and compared against baseline Generative AI, Generative AI + XAI, and Generative AI + CBR models. Results demonstrate that the proposed hybrid framework consistently outperforms existing approaches, achieving factual consistency up to 90.10%, explanation faithfulness of 89.50%, retrieval precision of 92.40%, and user trust scores of 88.60%. In addition, the framework reduces hallucination rates by more than 40% and significantly improves robustness under lexical, structural, and noisy perturbations. The findings confirm that integrating generative intelligence with case-based reasoning and explainability provides a scalable and trustworthy AI paradigm capable of continuous learning, transparent reasoning, and reliable real-world deployment across dynamic application domains.