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 accuses IG Metall member of secretly recording Giga Berlin meeting
The union has denied the electric vehicle maker’s allegations.
Police seized the computer of an IG Metall member at Tesla Giga Berlin on Tuesday amid allegations that a works council meeting was secretly recorded.
The union has denied the electric vehicle maker’s allegations.
In a post on X, Gigafactory Berlin plant manager André Thierig stated that an external union representative from IG Metall attended a works council meeting and allegedly recorded the session. Thierig described the event as “truly beyond words.”
“What has happened today at Giga Berlin is truly beyond words! An external union representative from IG Metall attended a works council meeting. For unknown reasons he recorded the internal meeting and was caught in action! We obviously called police and filed a criminal complaint!” Thierig wrote in his post on X.
Police later confirmed to local news outlet rbb24 that officers did seize a computer belonging to an IG Metall member at the Giga Berlin site on Tuesday afternoon. Tesla stated that employees had contacted authorities after discovering the alleged recording.
IG Metall denied Tesla’s accusations, arguing that its representative did not record the meeting. The union alleged that Tesla’s claim was simply a tactic ahead of upcoming works council elections.
The next works council election at Giga Berlin is scheduled for March 2 to 4, 2026. The facility’s management had confirmed the dates to local news outlets. The official announcement marks the start of the election process and campaign period.
Approximately 11,000 employees are eligible to participate in the vote.
The previous works council election at the plant took place in 2024, and it was triggered by a notable increase in workforce size. Under German labor law, regular works council elections must be held every four years between March 1 and May 31.
Elon Musk
Elon Musk’s xAI plants flag in Bellevue AI hotspot
The lease places xAI’s new office in one of the region’s fastest-growing tech hubs.
Elon Musk’s artificial intelligence company xAI has leased a full floor at Lincoln Square South in downtown Bellevue, WA, as per city permit filings.
The lease places xAI’s new office in one of the region’s fastest-growing tech hubs.
Public records indicate that xAI leased roughly 24,800 square feet in Lincoln Square South. The location was previously occupied by video game company Epic Games. Lincoln Square South is part of the Bellevue Collection, which is owned by Kemper Development Co.
The lease was first referenced in January by commercial real estate firm Broderick Group, which noted that an unnamed tenant had secured the space, as stated in a report from the Puget Sound Business Journal. Later filings identified xAI as the occupant for the space.
xAI has not publicly commented on the lease.
xAI hinted at plans to open an office in the Seattle area back in September, when the startup posted job openings with salaries ranging from $180,000 to $440,000. At the time, the company had narrowed its location search to cities on the Eastside but had not finalized a lease.
xAI’s Bellevue expansion comes as Musk continues consolidating his businesses. Last week, SpaceX acquired xAI in a deal that valued the artificial intelligence startup at $250 billion. SpaceX itself is now valued at roughly $1.25 trillion and is expected to pursue an initial public offering (IPO) later this year.
Musk already has a significant presence in the region through SpaceX, which employs about 2,000 workers locally. That initiative, however, is focused largely on Starlink satellite development.
Bellevue has increasingly become a center for artificial intelligence companies. OpenAI has expanded its local office footprint to nearly 300,000 square feet. Data infrastructure firms such as Crusoe and CoreWeave have also established offices downtown.
Elon Musk
SpaceX blocks unauthorized Starlink terminals used by Russian troops
Ukrainian officials confirmed that Starlink terminals believed to be used by Russian troops were disabled after coordination with SpaceX.
SpaceX has taken steps to block unauthorized use of its Starlink satellite internet network, a move Ukrainian officials stated is already disrupting Russian military communications.
Russian units lose a key communications tool
As per a report from The Guardian, Ukrainian defense officials have confirmed that Starlink terminals believed to be used by Russian troops were recently disabled after coordination with SpaceX. The move reportedly affected frontline communications and drone operations, especially in areas where traditional military radios are unreliable or easily jammed.
For months, Russian units had relied on large numbers of illicitly obtained Starlink terminals to stay connected along the front. The satellite internet service allowed faster coordination and more precise drone use for Russian forces.
Several Russian military bloggers close to frontline units have acknowledged the impact of the Starlink shutdown, with some describing sudden connectivity problems in the satellite internet service.
Russia lacks comparable replacement
Russia does not have a satellite internet system that matches Starlink’s speed, coverage, and ease of deployment. Alternatives such as fiber-optic lines, short-range wireless links, and digital radio systems take longer to install and work inadequately for fast-moving units.
Russia does operate limited satellite communications through state-linked providers, but those systems rely mainly on geostationary satellites, which are notably slower. Coverage is uneven, and data capacity is far lower than Starlink’s low-Earth-orbit network.
For now, Ukraine has stated that it has introduced a verification system that allows only approved Starlink terminals to connect. Devices believed to be linked to Russian forces are blocked from the network. That being said, Ukrainian officials have also claimed Russian units are trying to work around the restrictions by asking civilians to register Starlink terminals in their names.
Tesla accuses IG Metall member of secretly recording Giga Berlin meeting
Elon Musk’s xAI plants flag in Bellevue AI hotspot
