The Full Self-Driving (FSD) AI development team at Tesla (TSLA) is currently working on two major versions of the technology. The first version or branch is for the general Tesla owners, and the other is the Robotaxi Unsupervised FSD.
The general version of Tesla FSD is less capable of self-driving Tesla vehicles compared to the Robotaxi FSD, which has been lately proven by extensive testing by early access riders.
Tesla CEO Elon Musk took to his social media platform X to share some new and interesting information on these two FSD branches. According to Musk, Tesla will soon be adding improvements from the Robotaxi FSD to the general FSD branch, and it’s going to bring a significant improvement for Tesla owners having FSD (Supervised) in their cars.
He wrote on X:
Your Tesla self-driving capability will see a step change improvement as we integrate upgrades for the Austin robotaxi build into the general production release
Tesla Robotaxi AI system used in Austin, Texas, is also constantly learning and improving as it navigates the streets. Besides this, the tech & AI automaker is constantly improving its mapping of the geofenced area of the city of Austin.
Just recently, Tesla expanded its Robotaxi geofence, the map of which became a point of discussion because of its odd shape.
Elon Musk further explained that his Tesla AI team is validating if the Robotaxi improvements would cause any regressions for general Tesla cars. Because this branch of the software is run in a geofenced area and FSD (Supervised) general Teslas are being self-driven across the United States and Canada.
“We need to validate that improvements for Austin don’t cause regressions elsewhere,” Musk posted on X (Twitter).
However, Elon Musk didn’t confirm which version of the Tesla FSD (Supervised) will be getting the Robotaxi branch improvements. As of this writing, the latest version is FSD v13.2.9 for HW4/AI4-based Tesla cars and hasn’t seen a new update since May.
Tesla is also working to merge both the Supervised and Unsupervised branches of Full Self-Driving AI software for a single stack. However, this work still needs polishing, according to the Tesla CEO.
“We have a more advanced model in alpha stage that has ~4X the params, but still requires a lot of polishing,” Musk said in June.
In another X post, Musk said that the Q3 2025 timeline is looking good for merging both FSD Supervised and Unsupervised stacks.
Robotaxi Launch in San Francisco Bay Area (California)
In the same discussion thread on X, Tesla’s Elon Musk also confirmed that the launch of Robotaxi in the San Francisco Bay Area is delayed due to regulatory hurdles.
However, Musk also praised the regulatory authorities of California for being reasonable in their requirements for Tesla Robotaxi service in SF Bay Area.
A few months ago, Tesla announced that it’s going to begin the Robotaxi service from Austin and California. However, the California launch is still pending approval.
Musk didn’t provide a solid timeline for the Tesla Robotaxi service launch in California.
Stay tuned for constant Tesla updates, follow us on:
Google News | Flipboard | X (Twitter) | WhatsApp Channel | RSS (Feedly).
Related Tesla Robotaxi News
- Musk promises a $30K or less Cybercab delivery before 2027, Tesla (TSLA) produces the first unit at Giga Texas
- Tesla files for additional trademarks for the Cybercab in some US states: Cybercar, Cybervehicle, Robotaxi
- Tesla Cybercab butterfly door function spotted in the wild (video)
- Tesla begins Robotaxi service without safety monitors (Unsupervised FSD) in Austin, Texas (videos)
- Tesla Cybercab has so much that’s not obvious on the surface, says Elon Musk
- Tesla (TSLA) begins crash-testing of the Cybercab robotaxi, Cortex 2 construction update at Giga Texas
 CEO Elon Musk says that the Robotaxi Unsupervised FSD improvements will be integrated into FSD (Supervised) in an upcoming FSD update.){kind=link}