Remote Sensing datasets are critical in problems requiring earth observation (EO) and the 3D LiDAR point cloud is one such dataset. Typically, the LiDAR point cloud dataset constitutes 3D spatial, topological, and complex geometrical information which possesses immense potential in applications involving machine-understandable 3D perception. However, the automated knowledge discovery of these relations from raw point clouds is challenging. Also, LiDAR point clouds are compute-intensive to process, attributed to their big size and unstructured. This chapter emphasizes the need for 3D information mining and recent advancements in Geo-Artificial Intelligence (Geo-AI) based solutions for 3D data processing. It proposes an implementation of a scalable LiDAR information mining framework. It also addresses the issues of heterogeneity in LiDAR data acquisition and the need for interoperability in LiDAR data processing by discussing the 3D geosemantics standards. A case study is presented for developing a Knowledge Base Question-Answering (KBQA) framework for 3D LiDAR data building on the proposed information mining framework. The chapter introduces a comprehensive picture of the subject matter for readers and practitioners working across diverse application domains such as forest inventory monitoring, real-time asset management, rapid disaster assessment, 3D mapping and surveying, and autonomous mobility.