Abstract: Metamaterials are artificially synthesized materials whose structures are meticulously designed and precisely controlled to exhibit unique properties not found in natural materials. These functional metamaterials have significant applications in fields such as optics, electromagnetics, acoustics, and mechanics. However, traditional metamaterial de- sign often relies on human expertise, resulting in lengthy design cycles, a lack of flexibility, and limited tunability, which hinder the rapid design and mass production of metamaterials. Efficient design and performance tuning of metamaterials have become crucial research directions in computer-aided design, mechanical engineering, and materials science. In recent years, with the rapid development of artificial intelligence algorithms, the application of intelligent algorithms in metamaterial design has been expanding, demonstrating advantages such as efficient optimization, generation of diverse structures, and shortened design cycles. This paper first introduces the basic concepts and development history of meta- materials, and then discusses the application fields and design issues of metamaterials from the perspective of practical needs. The paper identifies the core issues of intelligent design algorithms as data representation and dataset construction, providing a detailed explanation and comparative analysis of these aspects. Additionally, it reviews the framework of intelligent optimization algorithms. Finally, the paper summarizes the challenges faced in the field of metamaterial design and forecasts future development trends.