Energetic materials have attracted significant attention due to their critical roles in national defense， aerospace， and specialized engineering applications. However， their research and development are hindered by high experimental costs， safety risks， and lengthy synthesis cycles， which greatly limit the rapid iteration and practical deployment of novel energetic compounds. In recent years， machine learning （ML） has emerged as a powerful tool in chemistry and materials science owing to its strong capabilities in data modeling and prediction. This review summarizes the latest advances in machine learning–assisted chemical synthesis， focusing on three major aspects： reaction prediction， synthesis route planning， and automated synthesis. Particular emphasis is placed on the potential value and limitations of applying ML techniques to energetic material synthesis. The key challenges—such as data scarcity and inconsistency， lack of safety evaluation frameworks， and limited experimental validation and model retraining—are also discussed. Finally， the review outlines future perspectives， including the establishment of standardized and shareable databases， and the development of high-throughput and automated experimental platforms tailored for energetic systems. This work aims to provide theoretical insights and methodological support for achieving efficient and intelligent synthesis of energetic materials.