Tesla FSD to Get Navigation Improvements and Pothole Detection

Tesla’s Full Self-Driving (Supervised) software is preparing to address two of the most persistent complaints among drivers: confusing navigation routing and the system’s inability to reliably avoid potholes.

What Happened

In a recent exchange on X, Shopify executive Farhan Thawar highlighted that while FSD v14.3.10 has made strides, broken asphalt and stubborn navigation routing remain significant friction points. Phil Duan, a principal software engineer on Tesla’s AI team, responded directly to confirm that "improvements on both are coming." Tesla has listed pothole avoidance as an upcoming improvement in release notes since FSD v14.3, though it was initially promised as far back as 2019. Current builds occasionally display glimpses of this behavior, with some testers recording vehicles centering over cracked asphalt, but the feature remains inconsistent.

Why It Matters

For daily commuters, these updates could significantly reduce the need for manual disengagement. The navigation overhaul aims to teach FSD to learn from driver habits by analyzing manual interventions, such as ignoring mapped detours for preferred highway exits. This builds on the recently launched Preferred Routes feature and will extend to parking, where the system will study preferences like reversing into driveways or hugging garage walls. Additionally, Tesla is working on integrating Grok into the FSD stack to allow natural voice instructions for routing and parking, further streamlining the autonomous driving experience.

The Bottom Line

While it remains unclear whether these upgrades will arrive in an imminent point release or debut with the upcoming FSD v15 architecture, Tesla is actively developing solutions for route logic and hazard detection to make autonomous commuting smoother and more predictable.