As the utilization of 3D visual sensors becomes more widespread, the volume of point cloud data continues to escalate, imposing significant demands on hardware storage and transmission. This surge in data has markedly influenced the processing speed and efficiency of 3D visual systems. As a result, there is a growing necessity for effective technologies in point cloud coding. The rise of deep learning has driven the development of numerous neural network-based coding methods, which have achieved remarkable success in image and video coding through end-to-end training optimization and more accurate entropy modeling. In recent years, this trend has gradually extended to point cloud coding, yielding extensive research advancements. However, deep learning-based point cloud coding methods face dual challenges. First, at the data level, point cloud data exhibits a massive scale compared to image and video, while deep learning models heavily rely on large-scale and high-quality training data. Second, at the methodological level, the diversity of deep learning coding approaches leads to a lack of unified standard framework and strategy, resulting in challenges for practical applications. Therefore, establishing standards is crucial for the development and application of deep learning-based point cloud coding. Furthermore, such standards can guide research directions to better promote the practical applications of point cloud coding technologies for data communications. This chapter provides a detailed discussion on AI-based point cloud coding standards, which have been investigated by the representative multimedia coding standardization groups during the past few years.