Tesla Autopilot is now enabling the car to perceive the space around it thanks to the development of its Occupancy Networks. Tesla’s Autopilot Software Director, Ashok Elluswamy, shared a detailed thread on Twitter about a recent workshop the Autopilot team held. He also shared the workshop on Twitter.
Presented some of the recent work from the Tesla Autopilot team at CVPR this year, especially about "Occupancy Networks" – our approach to solve general obstacle detection and using it to enable sophisticated collision avoidance. Full talk here: https://t.co/wpGXlNHaWl (1/12)
— Ashok Elluswamy (@aelluswamy) August 21, 2022
In the video and Twitter Thread, Ashok explained how Tesla developed Occupancy Networks to literally give the car a sense of its surroundings. Humans have the ability to understand the objects around them at any given time. Is that car down the road moving at a slow speed or a fast speed? Do I, a pedestrian, have enough time to get across the street before being hit? What is that in the middle of the road? What is that falling from the sky? I should move out the way.
These reactions to scenarios and split-second decisions come naturally to humans. Tesla’s Autopilot Team is working to program the vehicles to do the same thing and this will save lives. Imagine the car being able to correctly detect its surroundings while the driver isn’t even paying attention. An example is sudden unintended accelerations (SUA). Ashok pointed out that Autopilot prevents around 40 of these types of accidents daily.
The workshop was held in June at this year’s Conference on Computer Vision and Pattern Recognition (CVPR.) in New Orleans. Ashok explained that the team developed Occupancy Networks which enable the car to predict the volumetric occupancy of everything around it.
Ashok explained that the typical approaches such as image-space segmentation of free space or pixel-wise depth have many issues. The solution to those issues is Occupancy Networks.
In other words, Occupancy Networks enable the car to perceive the space around it and determine whether or not it can drive in that space. For example, if a UFO were to suddenly crash in front of you while you’re driving, you would react quickly in the safest way possible. This is what the Autopilot Team is training the software to do.
Ashok shared details of how Occupancy Networks used Neural Radience Fields (NeRFs). “The occupancy representation of these networks allows for differentiable rendering of images (based on the Neural Radiance Fields work). However, unlike typical NeRFs, which are per scene, these occupancy nets generalize across scenes.”
These predictions are already used to prevent a lot of collisions. For e.g., Autopilot prevents ~40 crashes / day where human drivers mistakenly press the accelerator at 100% instead of the brakes. In the video Autopilot automatically brakes, saving this person's legs (7/12) pic.twitter.com/XtMssPT9cM
— Ashok Elluswamy (@aelluswamy) August 21, 2022
You can read Ashok’s full Twitter thread here and you can watch his presentation here. We are a little over a month before Tesla’s AI Day and I’m sure Tesla will share more about the life-saving technology it is working on as well as the Optimus Bot.
Dr. Know It All recently published a video about the new 10.69 update and shared his thought about Occupancy Network.
In a message on Twitter, he told me, “The beauty of Occupancy Networks is that the car doesn’t have to know what the objects it sees are, it just has to know that they are there in order to avoid them!”
Note: Johnna is a Tesla shareholder and supports its mission.
Your feedback is important. If you have any comments, concerns, or see a typo, you can email me at johnna@teslarati.com. You can also reach me on Twitter @JohnnaCrider1
Elon Musk
Tesla to increase Full Self-Driving subscription price: here’s when
Tesla will increase its Full Self-Driving subscription price, meaning it will eventually be more than the current $99 per month price tag it has right now.
Already stating that the ability to purchase the suite outright will be removed, Tesla CEO Elon Musk said earlier this week that the Full Self-Driving subscription price would increase when its capabilities improve:
“I should also mention that the $99/month for supervised FSD will rise as FSD’s capabilities improve. The massive value jump is when you can be on your phone or sleeping for the entire ride (unsupervised FSD).”
This was an expected change, especially as Tesla has been hinting for some time that it is approaching a feature-complete version of Full Self-Driving that will no longer require driver supervision. However, with the increase, some are concerned that they may be priced out.
$99 per month is already a tough ask for some. While Full Self-Driving is definitely worth it just due to the capabilities, not every driver is ready to add potentially 50 percent to their car payment each month to have it.
While Tesla has not revealed any target price for FSD, it does seem that it will go up to at least $150.
I should also mention that the $99/month for supervised FSD will rise as FSD’s capabilities improve.
The massive value jump is when you can be on your phone or sleeping for the entire ride (unsupervised FSD). https://t.co/YDKhXN3aaG
— Elon Musk (@elonmusk) January 23, 2026
Additionally, the ability to purchase the suite outright is also being eliminated on February 14, which gives owners another reason to be slightly concerned about whether they will be able to afford to continue paying for Full Self-Driving in any capacity.
Some owners have requested a tiered program, which would allow people to pay for the capabilities they want at a discounted price.
Unsupervised FSD would be the most expensive, and although the company started removing Autopilot from some vehicles, it seems a Supervised FSD suite would still attract people to pay between $49 and $99 per month, as it is very useful.
Tesla will likely release pricing for the Unsupervised suite when it is available, but price increases could still come to the Supervised version as things improve.
This is not the first time Musk has hinted that the price would change with capability improvements, either. He’s been saying it for some time. In 2020, he even said the value of FSD would “probably be somewhere in excess of $100,000.”
The FSD price will continue to rise as the software gets closer to full self-driving capability with regulatory approval. It that point, the value of FSD is probably somewhere in excess of $100,000.
— Elon Musk (@elonmusk) May 18, 2020
News
Tesla starts removing outright Full Self-Driving purchase option at time of order
Tesla has chosen to axe the ability to purchase Full Self-Driving outright from a select group of cars just days after CEO Elon Musk announced the company had plans to eliminate that option in February.
The company is making a clear-cut stand that it will fully transition away from the ability to purchase the Full Self-Driving suite outright, a move that has brought differing opinions throughout the Tesla community.
Earlier this week, the company also announced that it will no longer allow buyers to purchase Full Self-Driving outright when ordering a pre-owned vehicle from inventory. Instead, that will be available for $99 per month, the same price that it costs for everyone else.
The ability to buy the suite for $8,000 for a one-time fee at the time of order has been removed:
NEWS: Tesla no longer allows buyers to purchase FSD outright in the U.S. when ordering a pre-owned vehicle directly from inventory. Tesla now gives you the option to either subscribe for $99/month, or purchase FSD outright after taking delivery (available until February 14th). pic.twitter.com/1xZ0BVG4JB
— Sawyer Merritt (@SawyerMerritt) January 23, 2026
This is a major move because it is the first time Tesla is eliminating the ability to purchase FSD outright for one flat fee to any of its vehicles, at least at the time of purchase.
It is trying to phase out the outright purchase option as much as it can, preparing people for the subscription-based service it will exclusively offer starting on February 14.
In less than a month, it won’t be available on any vehicle, which has truly driven some serious conversation from Tesla owners throughout the community.
There’s a conflict, because many believe that they will now lose the ability to buy FSD and not pay for it monthly, which is an attractive offer. However, others believe, despite paying $8,000 for FSD, that they will have to pay more money on top of that cost to get the unsupervised suite.
Additionally, CEO Elon Musk said that the FSD suite’s subscription price would increase over time as capabilities increase, which is understandable, but is also quite a conflict for those who spent thousands to have what was once promised to them, and now they may have to pay even more money.
News
Tesla Robotaxi has a highly-requested hardware feature not available on typical Model Ys
These camera washers are crucial for keeping the operation going, as they are the sole way Teslas operate autonomously. The cameras act as eyes for the car to drive, recognize speed limit and traffic signs, and travel safely.
Tesla Robotaxi has a highly-requested hardware feature that is not available on typical Model Ys that people like you and me bring home after we buy them. The feature is something that many have been wanting for years, especially after the company adopted a vision-only approach to self-driving.
After Tesla launched driverless Robotaxi rides to the public earlier this week in Austin, people have been traveling to the Lone Star State in an effort to hopefully snag a ride from one of the few vehicles in the fleet that are now no longer required to have Safety Monitors present.
BREAKING: Tesla launches public Robotaxi rides in Austin with no Safety Monitor
Although only a few of those completely driverless rides are available, there have been some new things seen on these cars that are additions from regular Model Ys, including the presence of one new feature: camera washers.
With the Model Y, there has been a front camera washer, but the other exterior “eyes” have been void of any solution for this. For now, owners are required to clean them manually.
In Austin, Tesla is doing things differently. It is now utilizing camera washers on the side repeater and rear bumper cameras, which will keep the cameras clean and keep operation as smooth and as uninterrupted as possible:
🚨 Tesla looks to have installed Camera Washers on the side repeater cameras on Robotaxis in Austin
pic.twitter.com/xemRtDtlRR— TESLARATI (@Teslarati) January 23, 2026
Rear Camera Washer on Tesla Robotaxi pic.twitter.com/P9hgGStHmV
— TESLARATI (@Teslarati) January 24, 2026
These camera washers are crucial for keeping the operation going, as they are the sole way Teslas operate autonomously. The cameras act as eyes for the car to drive, recognize speed limit and traffic signs, and travel safely.
This is the first time we are seeing them, so it seems as if Safety Monitors might have been responsible for keeping the lenses clean and unobstructed previously.
However, as Tesla transitions to a fully autonomous self-driving suite and Robotaxi expands to more vehicles in the Robotaxi fleet, it needed to find a way to clean the cameras without any manual intervention, at least for a short period, until they can return for interior and exterior washing.
