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

Elon Musk

Tesla is sending its humanoid Optimus robot to the Boston Marathon

Tesla’s Optimus robot is heading to the Boston Marathon finish line

Published

on

By

Tesla’s Optimus humanoid robot will be stationed at the Tesla showroom at 888 Boylston Street in Boston, right along the final stretch of the Boston Marathon today, ready to cheer on runners and pose for photos with spectators.

According to a Tesla email shared by content creator Sawyer Merritt on X, Optimus will be at the Boston Boylston Street showroom on April 20, coinciding with Marathon Monday weekend. The Boston Marathon finishes on Boylston Street, and the surrounding area draws hundreds of thousands of spectators along with international broadcast coverage. Placing Optimus there puts it in front of a massive public audience at zero advertising cost.

The Tesla showroom is at 888 Boylston Street, between Gloucester Street and Fairfield Street. The final mile of the marathon runs directly along Boylston Street, with runners passing the big stores before reaching the finish line at Copley Square.

Optimus was first announced at Tesla’s AI Day event on August 19, 2021, when Elon Musk presented a vision for a general-purpose robot designed to take on dangerous, repetitive, and unwanted tasks. In March 2026, Optimus appeared at the Appliance and Electronics World Expo in Shanghai, where on-site staff stated that mass production of the robot could begin by the end of 2026. Before that, it showed up at the Tesla Hollywood Diner opening in July 2025 and at a Miami showroom event in December 2025.

Tesla’s well-calculated display of Optimus gives the public a low-pressure first encounter with a robot that Tesla is preparing  to soon deploy at scale. The company has previously indicated plans to manufacture Optimus robots at its Fremont facility at up to 1 million units annually, with an Optimus production line at Gigafactory Texas targeting 10 million units per year.

Tesla showcases Optimus humanoid robot at AWE 2026 in Shanghai

Advertisement

Musk has said that Optimus “has the potential to be more significant than the vehicle business over time,” and separately that roughly 80 percent of Tesla’s future value will come from the robot program. Whether that holds depends on production execution. For now, Boston gets a preview of what that future looks like, standing at the finish line on Boylston Street while 32,000 runners pass by.

Continue Reading

News

Tesla expands Unsupervised Robotaxi service to two new cities

This expansion builds directly on Tesla’s existing operations. Robotaxi has been ramping unsupervised rides in Austin for months and maintains activity in the San Francisco Bay Area.

Published

on

Credit: Tesla

Tesla has taken a major step forward in its autonomous ride-hailing ambitions.

On April 18, the company’s official Robotaxi account announced that Robotaxi service is now rolling out in Dallas and Houston, Texas. The update signals the rapid scaling of unsupervised autonomous operations in the Lone Star State.

The announcement includes a compelling 14-second video captured from inside a Model Y. Shot from the passenger perspective, the footage shows the vehicle navigating suburban roads in both cities with zero driver intervention, with no Safety Monitor to be seen.

Tesla also shared geofence maps highlighting the initial service areas: a compact zone in Houston covering parts of Willowbrook and Jersey Village, and a similarly defined area in Dallas near Highland Park and central neighborhoods.

Advertisement

This expansion builds directly on Tesla’s existing operations. Robotaxi has been ramping unsupervised rides in Austin for months and maintains activity in the San Francisco Bay Area.

With Dallas and Houston now live, Texas hosts three active hubs—an impressive concentration that triples the company’s Lone Star footprint in just weeks. The move aligns with Tesla’s Q4 2025 earnings guidance, which outlined a broader H1 2026 rollout across seven U.S. cities, including Phoenix, Miami, Orlando, Tampa, and Las Vegas.

Texas offers favorable regulations, high ride-share demand, and relatively straightforward suburban-to-urban driving patterns ideal for early autonomous scaling. While initial geofences appear modest—roughly 25 square miles per city—Tesla has historically expanded these zones quickly as it gathers real-world data.

Tesla confirms Robotaxi expansion plans with new cities and aggressive timeline

Advertisement

Unsupervised operation marks a critical milestone: passengers can summon, ride, and exit without safety drivers, a leap beyond many competitors still requiring human oversight.

For Tesla, the implications are significant. Successful scaling in major metros could accelerate the transition to a fully driverless fleet, unlocking new revenue streams and validating years of Full Self-Driving investment.

Riders gain convenient, potentially lower-cost mobility, while the company edges closer to Elon Musk’s vision of Robotaxis transforming urban transport.

As Tesla pushes into more cities this year, today’s launch in Dallas and Houston underscores its momentum. Hopefully, Tesla will be able to expand unsupervised rides to another U.S. state soon, which will mark yet another chapter in this short-but-encouraging Robotaxi story.

Advertisement
Continue Reading

News

Tesla is pushing Robotaxi features to owner cars with Spring Update

Tesla has quietly begun rolling out one of its most forward-looking Robotaxi-inspired features to existing customer vehicles.

Published

on

Tesla is starting to push Robotaxi features to owner cars, and the first instances are coming as the Spring 2026 Update starts to roll out.

Tesla has quietly begun rolling out one of its most forward-looking Robotaxi-inspired features to existing customer vehicles.

With the 2026 Spring Update (version 2026.14+), the rear passenger display now features a fully interactive navigation map that works while the car is driving — a capability previously reserved for Tesla Robotaxi.

Until now, Tesla’s rear displays have been largely limited to media controls, climate settings, and static route overviews. The new interactive map transforms the backseat into an active navigation hub, exactly the kind of passenger-first interface Tesla has been prototyping for its driverless fleet.

In a Robotaxi, where no one sits behind the wheel, every rider will need intuitive, real-time map access. By shipping this UI into thousands of owner cars months ahead of the Cybercab’s planned unveiling, Tesla is stress-testing the software in real-world conditions and giving loyal customers an early taste of the autonomous future.

The rollout is still in its early wave. Only a small number of vehicles have received 2026.14.1 so far, but the feature is expected to expand rapidly in the coming weeks. Owners of Model S, Model X, Model 3, Model Y, and Cybertruck are all eligible.

Advertisement

For buyers of the new Signature Edition Model S and X Plaid vehicles — whose deliveries begin in May — the update will likely arrive shortly after they take delivery, meaning the final chapter of Tesla’s flagship lineup will ship with cutting-edge Robotaxi preview tech baked in.

Elon Musk has long emphasized that Tesla ships supporting infrastructure well before new products launch. This rear-map rollout is a textbook example of that philosophy — quietly preparing both the software and the customer base for a world of fully driverless rides.

While the interactive map may seem like a modest convenience upgrade on the surface, its deeper purpose is unmistakable. Tesla is using its massive installed base of vehicles as a proving ground for the exact passenger experience that will define the Robotaxi era.

For current owners, it’s a free preview of tomorrow’s mobility; for the company, it’s invaluable data and real-world validation before the Cybercab hits the streets.

Advertisement
Continue Reading