Connect with us
tesla v11 tesla v11

News

Tesla FSD Beta V11.4.1 has ‘major architectural improvements’: Elon Musk

Credit: Whole Mars Blog | Twitter

Published

on

Tesla’s release notes for FSD Beta V.11.4.1 are out now. 

Version 11.4.1 may seem like another minor update, but Elon Musk commented that it introduces some significant improvements to FSD Beta. In a tweet, Musk hinted about what’s to come in FSD Beta V.12. 

“As mentioned earlier, V11.4.1 has major architectural improvements it’s actually much more than a point release. Should arguably be v12.0, but that reserved for when FSD is fully AI from video in to control out,” Musk commented.

Below are all the improvements introduced in V11.4.1. 

Advertisement
  • Improved control through turns, and smoothness in general, by improving geometry, curvature, position, type and topology of lanes, lines, road edges, and restricted space. Among other improvements, the perception of lanes in city streets improved by 36%, forks improved by 44%, merges improved by 27% and turns improved by 16%, due to a bigger and cleaner training set and updated lane-guidance module. 
  • Added lane-guidance inputs to the Occupancy Network to improve detections of long-range roadway features, resulting in a 16% reduction in false negative median detections. 
  • Improved ego’s assertiveness for crossing pedestrians in cases where ego can easily and safely cross before the pedestrian. 
  • Improved motorbike recall by 8% and increased vehicle detection precision to reduce false positive detections. These models also add more robustness to variance in vision frame-rate.
  • Reduced interventions cause by other vehicles cutting into ego’s lane by 43%. This was accomplished by creating a framework to probabilistically anticipate objects that may cut into ego’s lane and proactively offset and/or adjust speed to position ego optimally for these futures. 
  • Improved cut-in control by reducing lane-centric velocity error by 40-50% for close-by vehicles. 
  • Improved recall for object partial lane encroachment by 20%, high yaw-rate cut-in by 40%, and cut-out by 26% by using additional features of the lane-change trajectory to improve supervision. 
  • Reduced highway false slowdowns related to underestimated velocities for faraway object by adding 68K videos to the training set with improved auto-labeled ground truth. 
  • Smoothed in-lane offsetting for large vehicles by tuning the amount of lateral jerk allowed for the maneuver. 
  • Improved lateral control for upcoming high-curvature merges to bias away from the merging lane. 

Do you notice a change in V11.4.1? I’d like to hear your thoughts on the FSD beta update. Please get in touch with me at maria@teslarati.com or via Twitter @Writer_01001101.

Maria--aka "M"-- is an experienced writer and book editor. She's written about several topics including health, tech, and politics. As a book editor, she's worked with authors who write Sci-Fi, Romance, and Dark Fantasy. M loves hearing from TESLARATI readers. If you have any tips or article ideas, contact her at maria@teslarati.com or via X, @Writer_01001101.

Advertisement
Comments

News

Tesla stuns with another FSD approval in Europe, its second in two days

Published

on

Tesla has stunned by gaining yet another approval for its Full Self-Driving suite in Europe, its second in two days and its fifth overall.

Belgium will be the latest country to allow Tesla owners to utilize FSD on public roads in Europe, joining a quickly growing list that started with the Netherlands, Lithuania, and Estonia.

On Tuesday, Denmark announced its approval of the FSD suite, which has now been followed by Belgium just one day later.

The country’s Minister of Mobility, Annick De Ridder, announced the approval on her X account, stating that she had just signed the approval of Tesla FSD. It now goes to the country’s homologation department for the last step of the approval process.

The Belgian approval is one of mighty importance because it truly shows how quickly countries in Europe could greenlight the FSD suite consecutively. Approvals are already coming in relatively quickly, which is a great sign.

Perhaps the next big development that could come from FSD approvals in Europe is an approval from a country like England, Italy, France, Spain, or Germany. It would be something to see how FSD would perform in a major European metro, such as London, Barcelona, Madrid, Paris, Rome, or Berlin.

Full Self-Driving does an excellent job of roaming around major U.S. cities like New York and Los Angeles, but other high-profile international cities of significance would truly mark a line in the sand for Tesla, which can simply enable any vehicle in its customer-owned fleet to run FSD with the correct approvals.

Continue Reading

Elon Musk

SpaceX’s Elon Musk relieves worries about orbital data centers

Published

on

Rendering of Elon Musk overlooking a Starship fleet (Credit: Grok)
Rendering of Elon Musk overlooking a Starship fleet (Credit: Grok)

SpaceX CEO Elon Musk recently confronted worries about orbital data centers and launching satellites in mass quantities in space, as some voiced concerns about crowding.

Musk’s SpaceX plans to combat the issue of needing data centers by launching them into space instead of taking up valuable real estate on Earth. It has been a major point of SpaceX’s future, including its looming IPO, which could be the largest ever.

In a recent interview filmed at SpaceX’s Starlink terminal factory in Bastrop, Texas, Elon Musk directly addressed concerns that deploying large numbers of AI satellites for orbital data centers could crowd Earth’s orbit. His message was straightforward and reassuring: space is vast beyond human intuition.

“Space is really big,” Musk said. “It’s not like space is gonna get crowded. Space is enormous. If you actually look at it relative to the Earth, the satellites are so tiny you can’t even see them.” He emphasized that even zooming in makes a satellite appear large, but from a planetary perspective, they are minuscule specks.

Musk pointed to SpaceX’s real-world experience operating roughly 10,000 Starlink satellites as evidence that large constellations can be managed safely. “We’ve got a pretty good idea of how to operate just really large constellations and do it safely,” he noted. SpaceX remains the only operator with meaningful experience at this scale, giving the company unique insight into tight orbital packing without compromising safety

The discussion highlighted SpaceX’s plans for “AI1” satellites—essentially orbiting racks of AI compute powered by massive solar arrays and cooled via radiative panels in space’s vacuum.

These satellites leverage proven Starlink V3 technology, making them simpler to design than communications satellites. A first-generation unit targets around 150 kW peak power, with a 70-meter wingspan for solar panels and radiators. Laser links will connect them to each other and the Starlink network, delivering low-latency access (on the order of a few milliseconds from low-Earth orbit).

FCC accepts SpaceX filing for 1 million orbital data center plan

Musk framed orbital data centers as a practical solution to Earth’s constraints on AI growth. Ground-based facilities face power shortages, water demands for cooling, and grid limitations. In space, constant sunlight (no day-night cycle), vacuum radiative cooling, and abundant solar energy offer clear advantages.

Production will ramp up at an expanded “Gigasat” factory in Bastrop, with solar manufacturing already underway and full AI satellite output expected at reasonable volume by the end of 2027. Starship’s rapid, high-volume launch capability, aiming for multiple flights per hour, will make massive deployment feasible.

Critics sometimes raise risks like space debris or Kessler syndrome, but Musk’s response underscores scale: even a million satellites would represent an imperceptible fraction of available orbital volume when viewed against Earth’s size. SpaceX’s automated collision avoidance and deorbiting designs for Starlink further mitigate concerns.

This vision ties into broader ambitions. Musk sees orbital AI compute as a step toward harnessing more of the Sun’s energy, advancing humanity on the Kardashev scale from a Type 0 civilization toward Type 1 and eventually Type 2. By moving power-hungry data centers off-planet, SpaceX aims to unlock orders-of-magnitude more compute while preserving Earth’s resources.

Musk’s comments should ease public anxiety. With proven operational expertise, incremental engineering, and the immensity of space itself, orbital data centers represent not overcrowding, but smart expansion into the final frontier.

Continue Reading

Investor's Corner

Tesla Full Self-Driving hits Level 4? One analyst says yes

Published

on

Credit: Tesla

Tesla Full Self-Driving (Supervised) is currently listed as a Level 2 suite in terms of its passenger cars. As its Robotaxi platform continues to move quickly, it has been recognized as a Level 4 ride-sharing program by the State of Texas, as Tesla recently self-certified itself.

However, a Wall Street analyst is arguing that Tesla (NASDAQ: TSLA) has effectively achieved Level 4 autonomy in most conditions in all of its vehicles, drawing on personal experience and data released by the company.

Alex Potter of Piper Sandler said in a note to investors on Wednesday that “Tesla has solved the self-driving puzzle,” pointing to decisions to offer insurance discounts for FSD-enabled policies as a signal of confidence, which is backed up by stellar safety records compared to human driving.

Investing.com initially reported on Potter’s new note.

Additionally, Potter looks at the recent start of Cybercab production at Giga Texas as a potential indication that Tesla is ready to offer some level of unsupervised driving at least in the near future. The Cybercab has no steering wheel or pedals, completely eliminating the ability for human input.

He also sees Tesla’s allocation of “several hundred million USD (if not $1B+)” as confidence internally, seeing as it would be tough to set aside that amount of capital toward a project that the company does not see as relatively near-term.

Forward thinking, especially as Cybercab has no human controls, it would make sense that Tesla is at least close to self-driving. How close is another question.

Tesla has routinely teased that unsupervised FSD is close, but there are still a lot of things it feels as if the company has to roll out some more capability, including unsupervised parking features, known as “Banish,” better operation with regional self-driving performance, and other improvements.

That is not to say that Tesla FSD is super impressive already. It has already completed coast-to-coast drives across the United States and Canada, it routinely takes the stress out of driving for most people, and it has proven through Tesla Safety Reports that it is safer and involved in accidents less frequently than humans.

Even Potter believes it is capable, as he used it to go from Missoula, Montana, to Minneapolis, Minnesota, back in April.

“There’s no substitute for personal experience,” he wrote.

Continue Reading