Germany is running one of the largest coordinated LiDAR campaigns in Europe. The Digital Twin Germany (DigiZ-DE) program, led by the Federal Agency for Cartography and Geodesy (BKG), is surveying 356,794 km² of the country at a point density of 40 pts/m², roughly four to eight times higher than typical national LiDAR surveys. The result is a unified 3D model of the entire country, supporting urban planning, crisis management, and long-term sustainability. At this scale, automation isn't a convenience. It's a necessity. In this session, Flai shares how AI-powered LiDAR point cloud classification is being applied across the DigiZ-DE program: classifying ground, vegetation, buildings, water, bridges, power lines, wind turbines, and project-specific object classes in a single automated run. Key takeaways: - How AI-driven LiDAR classification scales to country-level point cloud volumes - Strip-level noise classification, granular power line categorization, and dedicated handling of project-specific classes like wind turbines - Why a 12-class ASPRS-based scheme, extended with custom classes, works at national scale - Local vs. cloud processing trade-offs for large LiDAR programs This talk is aimed at national mapping agencies, regional surveying authorities, and geospatial production teams running large LiDAR programs. It offers a concrete look at what AI-supported classification makes possible today. Background: an independent overview was published in GIM International, "Processing nationwide data for Germany's digital twin": https://www.gim-international.com/content/article/processing-nationwide-data-for-germany-s-digital-twin