Tesla Confirms In-House AI Map for FSD v15 Rollout
Tesla announced during an earnings call that it is developing an in-house AI-based mapping and navigation system to replace third-party dependencies, which will roll out starting with FSD v15. This new architecture leverages Tesla's massive vehicle fleet to automatically collect street data, speed limits, traffic signs, and parking details. This shift marks a major architectural change in Tesla's autonomous driving stack, addressing ongoing FSD pain points like speed limit recognition, parking, and route planning. By cutting ties with external map providers, Tesla gains tighter integration between its driving software and map data to enable smoother and safer navigation. The new AI map will initially handle departure and arrival routing in FSD v15 before expanding to support parking preferences and voice-controlled routing over several months. It will also empower the system to detect and log road hazards like potholes, obstructed viewpoints, and dangerous intersections.
## BACKGROUND
Tesla Full Self-Driving (FSD) is an advanced driver-assistance system that provides semi-autonomous navigation and driving capabilities. Previously, Tesla relied on third-party APIs—such as Google Maps for base tiles and street data—combined with its own software to compute routes and display supercharger locations.