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 executive moves on after 13 years: ‘It has been a privilege to serve’
“It is challenging to encapsulate 13 years in a single post. The journey at Tesla has been one of continuous evolution. From the technical intricacies of designing, building, and operating one of the world’s largest AI clusters to impactful contributions in IT, Security, Sales, and Service, it has been a privilege to serve,” Jegannathan said in the post.
Tesla executive Raj Jegannathan is moving on from the company after 13 years, he announced on LinkedIn on Monday.
“It is challenging to encapsulate 13 years in a single post. The journey at Tesla has been one of continuous evolution. From the technical intricacies of designing, building, and operating one of the world’s largest AI clusters to impactful contributions in IT, Security, Sales, and Service, it has been a privilege to serve,” Jegannathan said in the post.
After starting as a Senior Staff Engineer in Fremont back in November 2012, Jegannathan slowly worked his way through the ranks at Tesla. His most recent role was Vice President of IT/AI Infrastructure, Business Apps, and Infosec.
However, it was reported last year that Jegannathan had taken on a new role, which was running the North American sales team following the departure of Troy Jones, who had held the position previously.
While Jegannathan’s LinkedIn does not mention this position specifically, it seemed to be accurate, considering Tesla had not explicitly promoted any other person to the role.
It is a big loss for Tesla, but not a destructive departure. Jegannathan was one of the few company executives who answered customer and fan questions on X, a unique part of the Tesla ownership experience.
Tesla to offer Full Self-Driving gifting program: here’s how it will work
It currently remains unclear if Jegannathan was removed from the position or if he left under his own accord.
“As I move on, I do so with a full heart and excitement for what lies ahead. Thank you, Tesla, for this wonderful opportunity!” he concluded.
The departure marks a continuing trend of executives leaving the company, as the past 24 months have seen some significant turnover at the executive level.
Tesla has shown persistently elevated executive turnover over the past two years, as names like Drew Baglino, Rohan Patel, Rebecca Tinucci, Daniel Ho, Omead Afshar, Milan Kovac, and Siddhant Awasthi have all been notable names to exit the company in the past two years.
There are several things that could contribute to this. Many skeptics will point to Elon Musk’s politics, but that is not necessarily the case.
Tesla is a difficult, but rewarding place to work. It is a company that requires a lot of commitment, and those who are halfway in might not choose to stick around. Sacrificing things like time with family might not outweigh the demands of Tesla and Musk.
Additionally, many of these executives have made a considerable amount of money thanks to stock packages the company offers to employees. While many might be looking for new opportunities, some might be interested in an early retirement.
Tesla is also in the process of transitioning away from its most notable division, automotive. While it still plans to manufacture cars in the millions, it is turning more focus toward robotics and autonomy, and these plans might not align with what some executives might want for themselves. There are a wide variety of factors in the decision to leave a job, so it is important not to immediately jump to controversy.
News
Lemonade launches Tesla FSD insurance program in Oregon
The program was announced by Lemonade co-founder Shai Wininger on social media platform X.
Tesla drivers in Oregon can now receive significant insurance discounts when using FSD, following the launch of Lemonade’s new Autonomous Car insurance program.
The program was announced by Lemonade co-founder Shai Wininger on social media platform X.
Lemonade launches FSD-based insurance in Oregon
In a post on X, Wininger confirmed that Lemondade’s Autonomous Car insurance product for Tesla is now live in Oregon. The program allows eligible Tesla owners to receive roughly 50% off insurance costs for every mile driven using Tesla’s FSD system.
“And… we’re ON. @Lemonade_Inc’s Autonomous Car for @Tesla FSD is now live in Oregon. Tesla drivers in Oregon can now get ~50% off their Tesla FSD-driven miles + the best car insurance experience in the US, bar none,” Wininger wrote in his post.
As per Lemonade on its official website, the program is built on Tesla’s safety data, which indicates that miles driven using FSD are approximately twice as safe as those driven manually. As a result, Lemonade prices those miles at a lower rate. The insurer noted that as FSD continues to improve, associated discounts could increase over time.
How Lemonade tracks FSD miles
Lemonade’s FSD discount works through a direct integration with Tesla vehicles, enabled only with a driver’s explicit permission. Once connected, the system distinguishes between miles driven manually and those driven using FSD, applying the discount automatically to qualifying miles.
There is no minimum FSD usage requirement. Drivers who use FSD occasionally still receive discounted rates for those miles, while non-FSD miles are billed at competitive standard rates. Lemonade also emphasized that coverage and claims handling remain unchanged regardless of whether a vehicle is operating under manual control or FSD at the time of an incident.
The program is currently available only to Teslas equipped with Hardware 4 or newer, running firmware version 2025.44.25.5 or later. Lemonade also allows policyholders to bundle Tesla insurance with renters, homeowners, pet, or life insurance policies for additional savings.
News
Tesla exec: Preparations underway but no firm timeline yet for FSD rollout in China
The information was related by Tesla China Vice President Grace Tao in a comment to local media.
Tesla has not set a specific launch date for Full Self-Driving in China, despite the company’s ongoing preparations for a local FSD rollout.
The information was related by Tesla China Vice President Grace Tao in a comment to local media.
Tesla China prepares FSD infrastructure
Speaking in a recent media interview, the executive confirmed that Tesla has established a local training center in China to support the full adaptation of FSD to domestic driving conditions, as noted in a report from Sina News. However, she also noted that the company does not have a specific date when FSD will officially roll out in China.
“We have set up a local training center in China specifically to handle this adaptation,” Tao said. “Once officially released, it will demonstrate a level of performance that is no less than, and may even surpass, that of local drivers.”
Tao also emphasized the rapid accumulation of data by Tesla’s FSD system, with the executive highlighting that Full Self-Driving has now accumulated more than 7.5 billion miles of real-world driving data worldwide.
Possible 2026 rollout
The Tesla executive’s comments come amidst Elon Musk’s previous comments suggesting that regulatory approval in China could arrive sometime this 2026. During Tesla’s annual shareholder meeting in November 2025, Musk clarified that FSD had only received “partial approval” in China, though full authorization could potentially arrive around February or March 2026.
Musk reiterated that timeline at the World Economic Forum in Davos, when he stated that FSD approval in China could come as early as February.
Tesla’s latest FSD software, version 14, is already being tested in more advanced deployments in the United States. The company has also started the rollout of its fully unsupervised Robotaxis in Austin, Texas, which no longer feature safety monitors.
Tesla executive moves on after 13 years: ‘It has been a privilege to serve’
Lemonade launches Tesla FSD insurance program in Oregon
