This study leverages recent advances in machine-based pattern recognition to estimate occurrence maps for over 600,000 species, using georeferenced data from the Global Biodiversity Information Facility (GBIF). A pilot application for priority-setting identifies 30 nations that host nearly 80 percent of threatened species with small ranges limited to a single country. The algorithms are designed for rapid map updates and estimating new maps as growth in GBIF species occurrence reports continues.