This paper reviews the latest advancements in artificial intelligence-assisted R&D project initiation, aiming to provide intelligent solutions for R&D management. It thoroughly examines the value of artificial intelligence technologies in four core areas: intelligent requirement analysis, technical feasibility assessment, market prospect forecasting, and automated risk identification. Furthermore, it proposes three forward-looking trends—enhanced intelligence, the establishment of industry standards, and deeper human-machine collaboration. These insights are expected to improve project approval success rates and shorten initiation timelines, driving a paradigm shift in R&D management from experience-based to data-driven decision-making. The review highlights how artificial intelligence, through machine learning, natural language processing, and data mining, effectively addresses chronic challenges in traditional initiation processes such as inefficiency, delayed decisions, and resource misallocation. It also identifies critical hurdles, including data quality, model interpretability, and organizational transformation, offering a vital reference framework for the future of intelligent R&D development.