Niantic's geospatial model is using geolocation data from scans players submit of real-world locations while playing the company's mobile games.

Pokemon Go developer Niantic reveals its long-term use of player data in the creation of a new large geospatial AI model.

The blog post explains that “Over the past five years, Niantic has focused on building our Visual Positioning System (VPS), which uses a single image from a phone to determine its position and orientation using a 3D map built from people scanning interesting locations in our games and Scaniverse,

The development of LGM builds upon Niantic’s existing Lightship Visual Positioning System (VPS), which has already mapped 10 million locations globally. These scans are uniquely valuable as they capture environments from a pedestrian perspective,

Data captured from Pokémon Go and other games are being used to create large AI models that can predict what buildings and other objects look like.

Niantic, the team behind Pokemon Go, is working on a new type of AI model that's training on data from its apps.

Pokemon Go” developer Niantic announced its new AI model on Nov. 12, created using data players have sent over the years. The large geospatial model (LGM), built using the San Francisco-based studio’s visual positioning system (VPS),