In recent years, with the rapid development of artificial intelligence (AI) technology, utilizing AI technology to improve 5G network performance has also become a research hotspot in the field of mobile communication. A large number of research results indicate that using AI technology can enhance multiple designs in existing 5G networks from multiple layers, thereby achieving the effect of improving system operation efficiency and user experience. Especially with the introduction of multiple wireless air interface technologies into AI based design, it will bring fundamental changes to wireless network design and be a key direction for future network design. The continuous integration of 5G and AI is the process of continuously building 5G intelligent dimensions. The intelligent dimension of 5G is a new resource dimension based on 5G big data and computing resources, and artificial intelligence technology. 5G intelligent dimension can be considered as a new dimension parallel to traditional wireless mobile communication in terms of time domain, frequency domain, and spatial domain. Compared to traditional dimensions, the construction of 5G intelligent dimensions requires three basic elements: 5G big data, computing resources, and artificial intelligence technology. 5G big data includes both raw data in 5G networks and data carried by 5G. Computing resources include 5G terminal computing power, 5G network element computing power, and cloud device computing power. Artificial intelligence technology is represented by machine learning. The construction of 5G intelligent maintenance can be defined as utilizing artificial intelligence technology, making reasonable use of 5G big data and computing resources, making 5G more intelligent and efficient, and applying it to intelligent adaptation of 5G networks to achieve high-quality and diverse services. Introducing multiple key technologies based on AI into practical networks still faces a series of critical issues that urgently need to be addressed, from basic theory to practical applications. These key issues include key use case selection, dataset construction, model management, and computational power support. For the mining of 5G big data, a series of AI basic theories and tools are needed. AI tools and algorithms have clear requirements for data and computing power. When designing an enhanced 5G network architecture, it is necessary to consider the data requirements and actual performance of different algorithms, as well as evaluate computing resources in combination with relevant algorithms. The powerful transmission capability of 5G networks will also drive the emergence of more AI based applications. The data collection, calculation, model deployment, and update methods of AI related applications need to be dynamically integrated with 5G networks to provide better services. At the beginning of the 5G international standard, AI based design was not introduced due to various factors. With the rapid development of AI technology and the continuous evolution of 5G international standards, designs that integrate with AI have gradually been considered. Since the R15 version, 5G wireless networks and core networks have gradually introduced AI based design, while also conducting a large amount of research on AI technology support. This article provides a detailed introduction and analysis of the continuous integration of 5G and AI based on the progress of international 5G standards. AI based air interface design is the core of building 5G intelligent dimensions. In 5G, AI based physical layer design involves the design of air interfaces and high-level layers. High level design based on AI started research before air ports, with a focus on use cases such as base station energy conservation and mobility management. The air interface design mainly focuses on the research and development of key technologies related to wireless AI