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.
Tesla has officially opened Robotaxi access to everyone and everyone, but there is one catch: you have to have an iPhone.
Tesla’s Robotaxi service in Austin and its ride-hailing service in the Bay Area were both officially launched to the public today, giving anyone using the iOS platform the ability to simply download the app and utilize it for a ride in either of those locations.
It has been in operation for several months: it launched in Austin in late June and in the Bay Area about a month later. In Austin, there is nobody in the driver’s seat unless the route takes you on the freeway.
In the Bay Area, there is someone in the driver’s seat at all times.
The platform was initially launched to those who were specifically invited to Austin to try it out.
Tesla confirms Robotaxi is heading to five new cities in the U.S.
Slowly, Tesla launched the platform to more people, hoping to expand the number of rides and get more valuable data on its performance in both regions to help local regulatory agencies relax some of the constraints that were placed on it.
Additionally, Tesla had its own in-house restrictions, like the presence of Safety Monitors in the vehicles. However, CEO Elon Musk has maintained that these monitors were present for safety reasons specifically, but revealed the plan was to remove them by the end of the year.
Now, Tesla is opening up Robotaxi to anyone who wants to try it, as many people reported today that they were able to access the app and immediately fetch a ride if they were in the area.
We also confirmed it ourselves, as it was shown that we could grab a ride in the Bay Area if we wanted to:
🚨 Tesla Robotaxi ride-hailing Service in Austin and the Bay Area has opened up for anyone on iOS
Go download the app and, if you’re in the area, hail a ride from Robotaxi pic.twitter.com/1CgzG0xk1J
— TESLARATI (@Teslarati) November 18, 2025
The launch of a more public Robotaxi network that allows anyone to access it seems to be a serious move of confidence by Tesla, as it is no longer confining the service to influencers who are handpicked by the company.
In the coming weeks, we expect Tesla to then rid these vehicles of the Safety Monitors as Musk predicted. If it can come through on that by the end of the year, the six-month period where Tesla went from launching Robotaxi to enabling driverless rides is incredibly impressive.
News
Tesla analyst sees Full Self-Driving adoption rates skyrocketing: here’s why
“You’ll see increased adoption as people are exposed to it. I’ve been behind the wheel of several of these and the different iterations of FSD, and it is getting better and better. It’s something when people experience it, they will be much more comfortable utilizing FSD and paying for it.”
Tesla analyst Stephen Gengaro of Stifel sees Full Self-Driving adoption rates skyrocketing, and he believes more and more people will commit to paying for the full suite or the subscription service after they try it.
Full Self-Driving is Tesla’s Level 2 advanced driver assistance suite (ADAS), and is one of the most robust on the market. Over time, the suite gets better as the company accumulates data from every mile driven by its fleet of vehicles, which has swelled to over five million cars sold.
The suite features a variety of advanced driving techniques that many others cannot do. It is not your typical Traffic-Aware Cruise Control (TACC) and Lane Keeping ADAS system. Instead, it can handle nearly every possible driving scenario out there.
It still requires the driver to pay attention and ultimately assume responsibility for the vehicle, but their hands are not required to be on the steering wheel.
It is overwhelmingly impressive, and as a personal user of the FSD suite on a daily basis, I have my complaints, but overall, there are very few things it does incorrectly.
Tesla Full Self-Driving (Supervised) v14.1.7 real-world drive and review
Gengaro, who increased his Tesla price target to $508 yesterday, said in an interview with CNBC that adoption rates of FSD will increase over the coming years as more people try it for themselves.
At first, it is tough to feel comfortable with your car literally driving you around. Then, it becomes second nature.
Gengaro said:
“You’ll see increased adoption as people are exposed to it. I’ve been behind the wheel of several of these and the different iterations of FSD, and it is getting better and better. It’s something when people experience it, they will be much more comfortable utilizing FSD and paying for it.”
Tesla Full Self-Driving take rates also have to increase as part of CEO Elon Musk’s recently approved compensation package, as one tranche requires ten million active subscriptions in order to win that portion of the package.
The company also said in the Q3 2025 Earnings Call in October that only 12 percent of the current ownership fleet are paid customers of Full Self-Driving, something the company wants to increase considerably moving forward.
News
Tesla scores major court win as judge rejects race bias class action
The ruling means the 2017 lawsuit cannot proceed as a class action because plaintiff attorneys were unable to secure testimony commitments from at least 200 workers.
Tesla scored a significant legal victory in California after a state judge reversed a class certification in a high-profile race harassment case involving 6,000 Black workers at its Fremont plant. The ruling means the 2017 lawsuit cannot proceed as a class action because plaintiff attorneys were unable to secure testimony commitments from at least 200 workers ahead of a 2026 trial, a threshold the judge viewed as necessary to reliably represent the full group.
No class action
In a late-Friday order, California Superior Court Judge Peter Borkon concluded that the suit could not remain a class action, stating he could not confidently apply the experiences of a much smaller group of testifying workers to thousands of potential class members. His ruling reverses a 2024 decision by a different judge who had certified the case under the belief that a trial of that size would be manageable, as noted in a Reuters report.
The lawsuit was originally filed by former assembly-line worker Marcus Vaughn, who alleged that Black employees at Tesla’s Fremont factory were exposed to various forms of racially hostile conduct, including slurs, graffiti, and instances of disturbing objects appearing in work areas. Tesla has previously said it does not tolerate harassment and has removed employees found responsible for misconduct. Neither Tesla nor the plaintiffs’ legal team immediately commented on the latest ruling.
Tesla’s legal challenges
While the decertification narrows the scope of this particular case, Tesla still faces additional litigation over similar allegations. A separate trial involving related claims brought by a California state civil rights agency is scheduled just two months after the now-vacated class trial date. The company is also contending with federal race discrimination claims filed by the U.S. Equal Employment Opportunity Commission, alongside several individual lawsuits it has already resolved.
For now, the reversal removes the large-scale exposure Tesla would have faced in a unified class trial, shifting the dispute back to individual claims rather than a single mass action. The case is Vaughn v. Tesla, filed in Alameda County Superior Court.
Tesla opens Robotaxi access to everyone — but there’s one catch
Tesla analyst sees Full Self-Driving adoption rates skyrocketing: here’s why
