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.
Elon Musk
Elon Musk’s xAI brings 1GW Colossus 2 AI training cluster online
Elon Musk shared his update in a recent post on social media platform X.
xAI has brought its Colossus 2 supercomputer online, making it the first gigawatt-scale AI training cluster in the world, and it’s about to get even bigger in a few months.
Elon Musk shared his update in a recent post on social media platform X.
Colossus 2 goes live
The Colossus 2 supercomputer, together with its predecessor, Colossus 1, are used by xAI to primarily train and refine the company’s Grok large language model. In a post on X, Musk stated that Colossus 2 is already operational, making it the first gigawatt training cluster in the world.
But what’s even more remarkable is that it would be upgraded to 1.5 GW of power in April. Even in its current iteration, however, the Colossus 2 supercomputer already exceeds the peak demand of San Francisco.
Commentary from users of the social media platform highlighted the speed of execution behind the project. Colossus 1 went from site preparation to full operation in 122 days, while Colossus 2 went live by crossing the 1-GW barrier and is targeting a total capacity of roughly 2 GW. This far exceeds the speed of xAI’s primary rivals.
Funding fuels rapid expansion
xAI’s Colossus 2 launch follows xAI’s recently closed, upsized $20 billion Series E funding round, which exceeded its initial $15 billion target. The company said the capital will be used to accelerate infrastructure scaling and AI product development.
The round attracted a broad group of investors, including Valor Equity Partners, Stepstone Group, Fidelity Management & Research Company, Qatar Investment Authority, MGX, and Baron Capital Group. Strategic partners NVIDIA and Cisco also continued their support, helping xAI build what it describes as the world’s largest GPU clusters.
xAI said the funding will accelerate its infrastructure buildout, enable rapid deployment of AI products to billions of users, and support research tied to its mission of understanding the universe. The company noted that its Colossus 1 and 2 systems now represent more than one million H100 GPU equivalents, alongside recent releases including the Grok 4 series, Grok Voice, and Grok Imagine. Training is also already underway for its next flagship model, Grok 5.
Elon Musk
Tesla AI5 chip nears completion, Elon Musk teases 9-month development cadence
The Tesla CEO shared his recent insights in a post on social media platform X.
Tesla’s next-generation AI5 chip is nearly complete, and work on its successor is already underway, as per a recent update from Elon Musk.
The Tesla CEO shared his recent insights in a post on social media platform X.
Musk details AI chip roadmap
In his post, Elon Musk stated that Tesla’s AI5 chip design is “almost done,” while AI6 has already entered early development. Musk added that Tesla plans to continue iterating rapidly, with AI7, AI8, AI9, and future generations targeting a nine-month design cycle.
He also noted that Tesla’s in-house chips could become the highest-volume AI processors in the world. Musk framed his update as a recruiting message, encouraging engineers to join Tesla’s AI and chip development teams.
Tesla community member Herbert Ong highlighted the strategic importance of the timeline, noting that faster chip cycles enable quicker learning, faster iteration, and a compounding advantage in AI and autonomy that becomes increasingly difficult for competitors to close.
AI5 manufacturing takes shape
Musk’s comments align with earlier reporting on AI5’s production plans. In December, it was reported that Samsung is preparing to manufacture Tesla’s AI5 chip, accelerating hiring for experienced engineers to support U.S. production and address complex foundry challenges.
Samsung is one of two suppliers selected for AI5, alongside TSMC. The companies are expected to produce different versions of the AI5 chip, with TSMC reportedly using a 3nm process and Samsung using a 2nm process.
Musk has previously stated that while different foundries translate chip designs into physical silicon in different ways, the goal is for both versions of the Tesla AI5 chip to operate identically. AI5 will succeed Tesla’s current AI4 hardware, formerly known as Hardware 4, and is expected to support the company’s Full Self-Driving system as well as other AI-driven efforts, including Optimus.
News
Tesla Model Y and Model 3 named safest vehicles tested by ANCAP in 2025
According to ANCAP in a press release, the Tesla Model Y achieved the highest overall weighted score of any vehicle assessed in 2025.
The Tesla Model Y recorded the highest overall safety score of any vehicle tested by ANCAP in 2025. The Tesla Model 3 also delivered strong results, reinforcing the automaker’s safety leadership in Australia and New Zealand.
According to ANCAP in a press release, the Tesla Model Y achieved the highest overall weighted score of any vehicle assessed in 2025. ANCAP’s 2025 tests evaluated vehicles across four key pillars: Adult Occupant Protection, Child Occupant Protection, Vulnerable Road User Protection, and Safety Assist technologies.
The Model Y posted consistently strong results in all four categories, distinguishing itself through a system-based safety approach that combines structural crash protection with advanced driver-assistance features such as autonomous emergency braking, lane support, and driver monitoring.
This marked the second time the Model Y has topped ANCAP’s annual safety rankings. The Model Y’s previous version was also ANCAP’s top performer in 2022.
The Tesla Model 3 also delivered a strong performance in ANCAP’s 2025 tests, contributing to Tesla’s broader safety presence across segments. Similar to the Model Y, the Model 3 also earned impressive scores across the ANCAP’s four pillars. This made the vehicle the top performer in the Medium Car category.
ANCAP Chief Executive Officer Carla Hoorweg stated that the results highlight a growing industry shift toward integrated safety design, with improvements in technologies such as autonomous emergency braking and lane support translating into meaningful real-world protection.
“ANCAP’s testing continues to reinforce a clear message: the safest vehicles are those designed with safety as a system, not a checklist. The top performers this year delivered consistent results across physical crash protection, crash avoidance and vulnerable road user safety, rather than relying on strength in a single area.
“We are also seeing increasing alignment between ANCAP’s test requirements and the safety technologies that genuinely matter on Australian and New Zealand roads. Improvements in autonomous emergency braking, lane support, and driver monitoring systems are translating into more robust protection,” Hoorweg said.
Elon Musk’s xAI brings 1GW Colossus 2 AI training cluster online
Tesla AI5 chip nears completion, Elon Musk teases 9-month development cadence
