Artificial Intelligence (AI) is reshaping the landscape of Evidence-Based Medicine (EBM) by enabling the dynamic integration of structured and unstructured data, including knowledge from Traditional Medicine (TM). Leveraging Natural Language Processing (NLP) and Machine Learning (ML), AI can extract, structure, and synthesize vast amounts of data from TM sources, enhancing their accessibility and relevance for researchers and clinicians. Large Language Models (LLMs) such as GPT-4 enable large-scale text analysis, knowledge extraction, and hypothesis generation, accelerating research and fostering a more holistic understanding of patient care. These models also improve access to culturally diverse perspectives and support more intuitive interactions in clinical settings. This presentation introduces TM-GPT, a research-oriented LLM for Traditional Medicine, developed through an integrated approach involving the acquisition of over 2 million documents and patents, advanced data preprocessing, model training, validation, and continuous learning. TM-GPT aims to enrich Real World Data (RWD) and promote evidence-informed integration of TM into global health systems, contributing to a more inclusive, data-driven future for healthcare.