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.”
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.”
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.
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.
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.
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.
Robotaxi now rolling out in Dallas & Houston 🤠 pic.twitter.com/G3KFQwqGxB
— Tesla Robotaxi (@robotaxi) April 18, 2026
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.
🚨 Tesla has expanded Robotaxi to two new cities: Houston and Dallas, joining Austin and the SF Bay Area as active Robotaxi areas https://t.co/S3Ck4EaGpR pic.twitter.com/N0qu0bcTyd
— TESLARATI (@Teslarati) April 18, 2026
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
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.
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.
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.
First look at Tesla’s v2026.14.1 Spring Update.
🧭Rear screen interactive map #teslaupdate #tesla #teslasrpingupdate pic.twitter.com/yH3T4U8qHp— Sergiu Mogan (@sergiumogan) April 17, 2026
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.
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.
News
Tesla Cybertruck sales bolstered by bold Musk move, report claims
If accurate, that means nearly one in every five Cybertrucks registered in the quarter was transferred internally within Musk’s business empire. The purchases, valued at more than $100 million, have continued into 2026.
A new report from Bloomberg claims Tesla Cybertruck sales were inflated by internal buyers, meaning companies owned by CEO Elon Musk, and most notably, SpaceX.
According to a new registration data analysis, a significant portion of the fourth quarter’s Cybertruck sales came from Musk companies.
In the fourth quarter of 2025, 7,071 Cybertrucks were registered in the United States. SpaceX, Musk’s rocket and satellite company, accounted for 1,279 of those vehicles—more than 18 percent of the total. Musk’s additional ventures, including xAI, the Boring Company, and Neuralink, acquired another 60 trucks during the same period.
Tesla Cybertruck just won a rare and elusive crash safety honor
If accurate, that means nearly one in every five Cybertrucks registered in the quarter was transferred internally within Musk’s business empire. The purchases, valued at more than $100 million, have continued into 2026.
These internal sales supplemented the Cybertruck’s overall performance for the quarter, as without them, sales would have plunged 51 percent. The vehicle, which has repeatedly been called “the best product Tesla has ever made,” has fallen short of expectations due to pricing.
When first unveiled back in 2019, Tesla had a $39,990, $49,990, and $69,990 configuration for sale. Those prices inflated significantly as the truck was not released to customers until 2023. Those who had placed orders for affordable configurations were priced out.
Sam Fiorani, VP of Global Vehicle Forecasting at AutoForecast Solutions, said, “Tesla is running out of buyers for the Cybertruck.” In reality, there are probably a lot of buyers, but they simply cannot afford the truck at its current price point.
The Cybertruck was supposed to broaden Tesla’s appeal beyond its core lineup of sleek sedans and SUVs. While it has done a lot for brand notoriety, it has not lived up to its monumental expectations, and it’s simply because the truck has not been as available as most had thought.
The truck is still the best-selling electric pickup in the country, outpacing rivals like the Ford F-150 Lightning and Chevrolet Silverado EV. It is also not uncommon for companies to use their own vehicles for internal operations, like Ford using its own Transit van for Mobile Service.
However, this much inventory of Cybertrucks being purchased by Musk’s companies is not what you love to see as a fan or investor.
Tesla expands Unsupervised Robotaxi service to two new cities
Tesla is pushing Robotaxi features to owner cars with Spring Update
