Connect with us

News

Tesla FSD v12 shifts away from ‘rules-based’ approach

Credit: @sabber_dev/Twitter

Published

on

Walter Isaacson’s Elon Musk biography is set to be published on Tuesday, and a new preview of the book illustrates details about Tesla’s development of the upcoming Full Self-Driving (FSD) version 12.

In an additional preview of his Musk biography for CNBC, Isaacson discusses the use of AI in the development of Tesla’s FSD v12, in a shift that took place within the last several months. Isaacson talks about Tesla’s recent development of the upcoming FSD v12, which he and Tesla demonstrate has moved away from a “rules-based” approach.

Notably, FSD v12 is expected to use billions of video frames from real-world driving incidents to train its neural network rather than using thousands of lines of code like previous versions. In a conversation with Musk last December, Tesla Autopilot employee Dhaval Shroff had likened the concept to the popular chatbot ChatGPT, instead for use with driving.

“It’s like ChatGPT, but for cars,” Shroff said. “We process an enormous amount of data on how real human drivers acted in a complex driving situation, and then we train a computer’s neural network to mimic that.”

Advertisement

Surprisingly enough, Tesla only shifted toward this “neural network planner” approach recently. By the beginning of this year, however, the neural network had already analyzed 10 million video clips based on the best-case-scenario drivers the system had access to. Musk instructed employees at the company’s Buffalo, New York facility who were in charge of analyzing the footage to train the AI on things “a five-star Uber driver would do.”

Moving from a rules-based to a network-path-based AI approach allowed FSD to use human driving data to avoid obstacles, even if breaking some rules was necessary. Shroff helped demonstrate the idea to Musk with a demo featuring trash bins, debris, and upturned traffic cones, which the car handled surprisingly well.

“Here’s what happens when we move from rules-based to network-path-based,” Shroff explained. “The car will never get into a collision if you turn this thing on, even in unstructured environments.”

Tesla FSD’s “autosteer on city streets” now available

Advertisement

Musk quickly took to the idea, as can be seen in a recent livestream of Tesla’s FSD v12 software in Palo Alto with Autopilot software director Ashok Elluswamy. He has repeatedly spoken about the upcoming software version’s impressive driving results, despite one small moment in the drive where the car almost ran a red light.

In any case, Musk could argue that the red-light moment is a good case for the need for self-driving software to continually learn. Given that it will constantly be trained from the video data generated by camera footage from real-world drivers, it should theoretically make it safer over time, according to Musk.

During development, Musk also reportedly latched onto the fact that it took over a million video clips for the neural network to begin performing well, though he looks forward to what significantly more data will do for FSD.

Still, critics and regulators have expressed concerns about the faults of human drivers training AI-based driving systems, and Tesla has repeatedly been questioned by the National Highway Traffic Safety Administration (NHTSA) about its Autopilot and FSD beta systems.

Advertisement

According to Isaacson, Tesla plans to release FSD v12 as soon as regulators approve it. Meanwhile, an ongoing study by the National Highway Safety Board is looking to determine if self-driving cars should be permitted to imitate human driving actions that blur traffic rules, such as creeping up at stop signs.

Musk said in April that he expects Tesla to reach full autonomy within a year, though he has also been known to share ambitious targets for the software in the past.

You can read Walter Isaacson’s full account of the development of Tesla FSD v12 here, in a CNBC preview of the upcoming Elon Musk biography.

What are your thoughts? Let me know at zach@teslarati.com, find me on X at @zacharyvisconti, or send your tips to us at tips@teslarati.com.

Advertisement

Zach is a renewable energy reporter who has been covering electric vehicles since 2020. He grew up in Fremont, California, and he currently lives in Colorado. His work has appeared in the Chicago Tribune, KRON4 San Francisco, FOX31 Denver, InsideEVs, CleanTechnica, and many other publications. When he isn't covering Tesla or other EV companies, you can find him writing and performing music, drinking a good cup of coffee, or hanging out with his cats, Banks and Freddie. Reach out at zach@teslarati.com, find him on X at @zacharyvisconti, or send us tips at tips@teslarati.com.

Advertisement
Comments

News

Tesla Full Self-Driving shows stunning maneuver in Europe to silence skeptics

In a striking demonstration of autonomous driving prowess, Tesla’s Full Self-Driving (FSD) system recently showcased its capabilities on the narrow rural roads of the Netherlands. Captured in two in-car videos, the system encountered scenarios that would challenge even the most experienced human drivers.

Published

on

Credit: Tesla

Tesla Full Self-Driving, fresh on the heels of its approval for operation on European roads for the first time, showed off a stunning maneuver that will certainly silence any skeptics on the continent.

Fresh off its approval in the Netherlands, Full Self-Driving is working toward a significant expansion into more parts of Europe.

In a striking demonstration of autonomous driving prowess, Tesla’s Full Self-Driving (FSD) system recently showcased its capabilities on the narrow rural roads of the Netherlands. Captured in two in-car videos, the system encountered scenarios that would challenge even the most experienced human drivers.

In the first clip, a wide tractor occupied more than half the lane on a tight two-way road. Rather than braking abruptly or forcing a collision risk, FSD smoothly edged the vehicle onto the adjacent bike path—using the extra space with precision—before seamlessly returning to the lane once clear.

Advertisement

The second clip was equally demanding: while overtaking a group of cyclists, an oncoming car approached at speed.

FSD maintained a safe, minimal buffer to the cyclists while timing the pass perfectly, avoiding any swerve or hesitation that could unsettle passengers or other road users.

Advertisement

This maneuver highlights FSD’s advanced spatial reasoning and predictive planning. On roads often under three meters wide, with no room for error, the system calculated available clearance in real time, incorporated shoulder and path geometry, and executed a controlled deviation without compromising safety.

It treated the bike path as a legitimate extension of navigable space, something many drivers might hesitate to do, while respecting Dutch road norms and cyclist priority.

Such feats align closely with a growing library of impressive FSD maneuvers documented on camera worldwide.

In urban Amsterdam, for instance, FSD has navigated the world’s densest cyclist environments, weaving through hundreds of unpredictable bike movements on canal-side streets with tram tracks and pedestrians.

Advertisement

One uncut drive showed it yielding smoothly at crossings, overtaking where needed, and even handling a near-perfect auto-park in a tight residential spot, demonstrating the same low-speed precision seen in the rural clips.

Teslas using FSD have tackled turbo roundabouts in the Netherlands, complex multi-lane circles notorious for geometry challenges, merging confidently while yielding to traffic. Similar clips depict smooth handling of construction zones, emergency vehicle pull-overs, and gated parking barriers, where the car stops precisely, waits for clearance, and proceeds without driver input.

Collectively, these examples illustrate FSD’s evolution toward handling the unpredictable.

The rural Netherlands maneuvers aren’t isolated. Instead, they reflect a pattern of spatial awareness, cyclist deference, and traffic anticipation seen from city streets to highways.

Advertisement

As FSD continues refining through real-world data, videos like this one are certainly building a compelling case for its readiness on Europe’s varied roads.

Continue Reading

News

Tesla utilizes its ‘Rave Cave’ for new awesome safety feature

Part of the massive interior overhaul of both the Model 3 “Highland” and Model Y “Juniper” was the addition of interior accent lighting to help bring out the mood of the vehicle, increase the customization of the interior, and to create a unique listening experience.

Published

on

Credit: Tesla | X

Tesla is utilizing its ‘Rave Cave’ for an awesome new safety feature that will arrive with the upcoming Spring Update for 2026.

Part of the massive interior overhaul of both the Model 3 “Highland” and Model Y “Juniper” was the addition of interior accent lighting to help bring out the mood of the vehicle, increase the customization of the interior, and to create a unique listening experience.

Tesla added a Sync Lights feature that will strobe the accent strips with the beat of the music.

It is one of the most unique and one of the coolest non-functional features of a Tesla, as it does not improve the driving of the vehicle, but makes it a cool and personal addition to the interior.

Advertisement

However, Tesla is going to take it one step further, as the Rave Cave lights will now be used for blind spot recognition. This feature will be added as the Spring 2026 Update starts to roll out.

Advertisement

Tesla writes:

“Accent lights now turn red when an object is in your blind spot and your turn signal is engaged, or when an approaching object is detected while parked.”

This neat new safety feature will now increase the likelihood of a driver, who is operating their Tesla manually, of seeing the blind spot warnings that are currently available on the A pillar and on the center touchscreen.

These new alerts will now warn drivers of cross traffic as they back out of a parking space with little to no visibility of what is coming. It is a great new addition that will only increase the safety of the vehicles, while also utilizing something that is already installed in these specific Model 3 and Model Y units.

Advertisement

The Model 3 and Model Y were the central focus of the Spring 2026 Update, especially considering the fact that the Model S and Model X are basically gone, with only a few hundred units left. Additionally, Tesla included new Immersive Sound and Car Visualization for the Model 3 and Model Y specifically in this new update.

Continue Reading

News

Tesla parked 50+ Cybercabs outside its Texas Factory with some crash tested

Dozens of Tesla Cybercabs have been spotted at Giga Texas crash testing facility ahead of launch.

Published

on

By

Tesla Cybercab fleet spotted at Gigafactory Texas [Credit: Joe Tegtmeyer)
Tesla Cybercab fleet spotted at Gigafactory Texas on April 13, 2026 [Credit: Joe Tegtmeyer)

Drone footage captured by longtime Giga Texas observer Joe Tegtmeyer shows over 50 units of Tesla Cybercab at the Austin factory campus, including several units clustered by Tesla’s on-site crash testing facility.

The outbound lot at Gigafactory Texas sits just outside the factory exit and serves as the primary staging area where finished vehicles are held before being loaded onto transport carriers or dispatched for validation testing. On any given day, the lot holds a mix of Model Y and Cybertruck units alongside the growing Tesla Cybercab fleet, as can be seen in the drone footage captured by Joe Tegtmeyer.

Tesla Cybercab fleet spotted at Gigafactory Texas [Credit: Joe Tegtmeyer)

Tesla Cybercab fleet spotted at Gigafactory Texas on April 13, 2026 [Credit: Joe Tegtmeyer)

Roughly 50 Cybercab units are visible across the campus, parked in tight organized rows. Most of the units visible still carry steering wheels and pedals, temporary additions Tesla included to satisfy current safety regulations while the vehicles accumulate real-world data ahead of full regulatory approval for a steering wheel-free design.

Tesla Cybercab fleet spotted at Gigafactory Texas [Credit: Joe Tegtmeyer)

Tesla Cybercab fleet spotted at Gigafactory Texas [Credit: Joe Tegtmeyer)

Tesla operates dedicated Crash Labs at both its Giga Texas and Fremont facilities that are purpose-built for controlled structural crash tests. Historically, automakers begin intensive crash testing roughly one to two months before volume production kicks off. The Cybertruck followed almost exactly that pattern. The Cybercab appears to be on the same track facility that we first saw back in October 2025.

Tesla Cybercab crash test units spotted at Gigafactory Texas [Credit: Joe Tegtmeyer)

Tesla Cybercab crash test units spotted at Gigafactory Texas [Credit: Joe Tegtmeyer)

The first production Cybercab rolled off the Giga Texas line on February 17, 2026. Volume production is now targeted for April. Musk previously wrote on X that “the early production rate will be agonizingly slow, but eventually end up being insanely fast,” and separately stated Tesla is targeting at least 2 million Cybercab units per year. Commercial robotaxi service in Austin is targeted for late 2026.

 

Advertisement
Continue Reading