For years, the progress of Tesla’s FSD has been measured by smoother turns, better lane centering, and more confident unprotected left turns. But as the system matures, a new, more subtle form of intelligence is emerging - one that shifts its attention to the human nuances of navigating roads. A new video posted to X shows the most recent FSD build, V13.2.9, demonstrating this in a remarkable real-world scenario.
Toll Booth Magic
In the video, a Model Y running FSD pulls up to a toll booth and smoothly comes to a stop, allowing the driver to handle payment. The car waits patiently as the driver interacts with the attendant. Then, at the precise moment the toll booth operator finishes the transaction and says “Have a great day”, the vehicle starts moving, proceeding through the booth - all without any input from the driver.
If you notice, there’s no gate here at this toll booth. This interaction all happened naturally with FSD.
While the timing was perfect, the FSD wasn’t listening to the conversation for clues (maybe one day, with Grok?) The reality, as explained by Ashok Elluswamy, Tesla’s VP of AI, is even more impressive.
It can see the transaction happening using the repeater & pillar cameras. Hence FSD proceeds on its own when the transaction is complete 😎
FSD is simply using the cameras on the side of the vehicle to watch the exchange between the driver and attendant. The neural network has been trained on enough data that it can visually recognize the conclusion of a transaction - the exchange of money or a card and the hands pulling away - and understands that this is the trigger to proceed.
The Bigger Picture
This capability is far more significant than just a simple party trick. FSD is gaining the ability to perceive and navigate a world built for humans in the most human-like fashion possible.
If FSD can learn what a completed toll transaction looks like, it’s an example of the countless other complex scenarios it’ll be able to handle in the future. This same visual understanding could be applied to navigating a fast-food drive-thru, interacting with a parking garage attendant, passing through a security checkpoint, or boarding a ferry or vehicle train — all things we thought that would come much later.
These human-focused interactions will eventually become even more useful, as FSD becomes ever more confident in responding to humans on the road, like when a police officer tells a vehicle to go a certain direction, or a construction worker flags you through a site. These are real-world events that happen every day, and it isn’t surprising to see FSD picking up on the subtleties and nuances of human interaction.
This isn’t a pre-programmed feature for a specific toll booth. It is an emergent capability of the end-to-end AI neural nets. By learning from millions of videos across billions of miles, FSD is beginning to build a true contextual understanding of the world. The best part - with a 10x context increase on its way, this understanding will grow rapidly and become far more powerful.
These small, subtle moments of intelligence are the necessary steps to a truly robust autonomous system that can handle the messy, unpredictable nature of human society.
Subscribe
Subscribe to our newsletter to stay up to date on the latest Tesla news, upcoming features and software updates.
Tesla has expanded its accessory lineup once again, this time with a new, sportier option for interior customization. Owners of the refreshed Model 3 (2024+) can now purchase a Dashboard Décor Wing in Carbon Fiber directly from Tesla in North America, for $300 USD ($415 CAD).
Update: This product is available in several regions, including North America, parts of Europe, and Australia. If it’s not available in your region yet, we expect Tesla to add it over the coming weeks.
A Sleek, Modern Look
This new trim piece replaces the standard textile dash with a high-quality carbon fiber accent, giving the minimalist interior a more aggressive and performance-oriented aesthetic.
According to Tesla’s official description, the part is dark gray and black carbon fiber that is cross-woven and finished with epoxy. That means that it won’t just look good - it’ll be resilient to whatever is left on your dash, or to the sun and heat.
Installation Included
A benefit of this official accessory is the all-inclusive price. The pricing includes not only the carbon fiber dashboard wing but also covers shipping to your preferred Service Center, where a professional installation of the new wing will be completed for you. This provides a seamless, factory-quality upgrade without the hassle or fitment issues of third-party alternatives.
Based on the service manuals, the replacement is expected to take approximately 90 minutes, meaning it should be a relatively quick visit to install it once it arrives at your local service center.
This new carbon fiber decor is the second official dashboard option that Tesla has offered for the refreshed Model 3. It joins the previously released Alcantara Décor Wing (also available for the Model Y), which adds a softer and cozier feel. Owners now have a distinct choice directly from Tesla: a plush and premium Alcantara, or a sleek and sporty Carbon Fiber.
Tesla’s recent Q2 2025 Earnings Call provided one of the clearest roadmaps for FSD as of late. While not as detailed as Tesla’s roadmaps from late 2024, these updates give us an idea of Tesla’s future plans for FSD.
Tesla’s comments on FSD during their Q&A session covered everything from a major upcoming FSD update to the latest status on introducing FSD to new territories and when Tesla expects FSD Unsupervised to be ready for customers.
FSD Improvements, Reduced Nag
The biggest news for owners is that a major update is on the way. The learnings and improvements from the Robotaxi FSD builds that power the network in Austin are being integrated into the main consumer FSD Supervised branch for customer vehicles. According to Elon, this will result in a “step change improvement” in capability.
The release involves carefully validating the improvements honed for Austin, so that they don’t cause regressions when deployed more broadly.
Elon once again touched on safety and supervision, a.k.a. the “nag.” Tesla’s teams have noted that many drivers disengage FSD in order to fiddle with controls—whether changing a song, changing their destination, or just glancing at their phone. They will then re-engage FSD once they are done, all to avoid an FSD strike. This is significantly less safe than just leaving FSD on the entire time while they look away. Musk discussed this before, but says there will be additional improvements. Tesla is working to reduce the nag and increase the window in which users can look away from the road. This improvement is expected to be included in the next build.
There’s more than just new capabilities and better controls on their way for the fleet, though. Elon mentioned an up to 10x improvement in parameters. Tesla previously said they were working on 4-4.5x the parameter count, but this newer number is far bigger. The challenge they are beginning to encounter is the limited available memory space for FSD on AI4 computers. The model needs to run in an optimized, predefined hardware package, and Tesla must trim their training models to fit within that package.
So far, they’ve been more than capable - and the indication of a 10x parameter increase is a doubling of what Tesla previously thought was possible. With a parameter increase that’s this big, FSD will be far more capable of handling edge cases, while also reacting much more human-like than ever before.
Release Timelines
During the call, Tesla indicated that a customer build is planned to be released sometime in Q3 2025, which would put us roughly in the July to September timeframe. They are actively working on this build at this point, and the key here is testing and validation, which could speed up or delay this newer FSD build.
That means Tesla will need to use its validation fleets (not the Robotaxi ones, but its other engineering vehicles) to validate the latest builds of FSD in various challenging areas across the country, ensuring there are no regressions in capability.
Once that is done, the build will be sent out to employees and the limited number of Wave 1 Early Access testers. If both groups green-light the release, it will begin a roll-out to the usual major Tesla influencers, and then we’ll see it begin rolling out to customers following their feedback.
Assuming that Tesla is validating and regression testing today, we can expect a public rollout around September, but if they’re still training the customer build, then it could be even later.
FSD Unsupervised
The Austin Robotaxi program is serving as the proving ground for FSD (Unsupervised), which will eventually allow for a truly driverless experience. The successful launch and rapid expansion of the Austin Network are the key steps that Tesla is taking to validate the safety and reliability of the system.
Tesla stated that it expects to launch FSD Unsupervised to certain US cities by the end of 2025. The key points here are select US cities, although expect other stipulations as well. While the hardware used in the Model Y Robotaxis is the same as other models (except for that second communication unit), not all models are trained equally. There’s a reason Tesla is using the Model Y for their Robotaxis — it performs notably better on FSD than other models.
Tesla closely validates its Robotaxi geofenced area. So the select cities may very well be those that have been validated and in which the Robotaxi already operates. Tesla could also limit it to the Model Y initially and even block off tricky intersections, much like they do with the Robotaxi.
Even then, getting FSD Unsupervised in customers’ hands will be a major milestone for Tesla, even if it’s very limited initially.
Global Rollout
The call also provided a fairly concise status update on Robotaxi and FSD expansion into other key markets, making it clear that the next steps aren’t on Tesla conducting country-specific training, but instead regulatory hurdles.
For Europe, Tesla is still working with the Netherlands for regulatory approval. Once their approval is secured, they’ll then need to obtain approval from the EU and the UNECE. This is a lengthy, multi-step approval process to secure an exemption for Tesla from the UNECE AV regulations. Elon is hopeful and confident that it could be approved as soon as the end of Q3, if not by the end of 2025.
In China, FSD is currently limited to AI4 vehicles, and Tesla is working on FSD for HW3 vehicles. The challenge in China is that each update requires regulatory approval and testing on each vehicle, making regulatory approvals a time-consuming process with each update.
Wrapping Up
After months without updates, it’s great to hear Tesla’s plans for FSD. It appears that AI4 owners can expect a major update in a couple of months, while Tesla continues to address regulatory issues in Europe and China. We also received our first concrete piece of information on when Tesla plans to roll out FSD Unsupervised to customers. This will likely be a much more gradual process than Tesla suggests during their earnings call, but each step brings us closer to true autonomy.
