Tesla Developing Proprietary AI-Based Maps for FSD V15

Tesla is developing its own AI-based mapping and navigation system, a move announced by Vice President of Autopilot Ashok Elluswamy during the company’s Q3 2026 earnings call. The new architecture is expected to debut with Full Self-Driving (FSD) v15, aiming to address specific limitations in current navigation and autonomy features.

What Happened

Currently, Tesla’s navigation relies on third-party content supplied by Google, including map tiles, street-level data, satellite imagery, points of interest, and traffic information, while Tesla handles route calculation and app development. During the earnings call, Elluswamy acknowledged shareholder concerns regarding FSD’s performance with speed-limit recognition, parking, and route selection. He stated, "You’re totally right that we shouldn’t be depending on external people’s maps," confirming that the new system will be "essentially AI-based mapping and navigation."

The new maps will utilize data collected by Tesla’s fleet to annotate streets with details such as traveling speed, stop signs, traffic lights, and parking lot configurations. This approach is designed to resolve existing FSD issues and enhance safety by identifying hazards like potholes, blind driveways, and dangerous intersections. Additionally, the system will support personalized routing and voice-controlled navigation customization.

Why It Matters

This shift marks a significant departure from Tesla’s reliance on external mapping providers, potentially allowing for tighter integration between navigation data and autonomy decisions. By using fleet data to map specific road conditions, Tesla aims to improve the accuracy of FSD in complex environments, particularly regarding speed limits and parking maneuvers. The update also introduces enhanced user customization for parking preferences, such as choosing spots based on proximity to other cars or handicap status.

The rollout of AI-based maps coincides with other navigation improvements, including the recently introduced Preferred Routes feature, which prioritizes previously driven paths. These developments suggest Tesla is focusing on refining the user experience and safety metrics of its autonomy suite by grounding its navigation in real-world, proprietary data rather than generic third-party inputs.

The Bottom Line

Tesla’s new AI-based maps are planned to begin rolling out with FSD v15, initially handling trip departures and arrivals. Elluswamy expects the system to expand into remaining navigation functions, including parking and voice control, over the following several months.