Elon Musk is hyping an upcoming Tesla ‘Full Self-Driving’ (FSD) update by claiming that it will have “less nag” – a far cry from its long-unfulfilled promise of unsupervised self-driving.
For the last few weeks, Tesla CEO Elon Musk has been hyping a new FSD update that he claims will include “10x higher parameters.”
We already reported how FSD owners should manage their expectations based on the last time Musk claimed a significant increase in parameter count.
Other than the increase in parameter count and the timing for the “end of September”, the CEO didn’t elaborate much on the update.
He again commented on it this weekend:
The FSD release in about 6 weeks will be a dramatic gain with a 10X higher parameter count and many other improvements. It’s going through training & testing now.
But this time he also added that the update should result in “much less nag”:
Once we confirm real-world safety of FSD 14, which we think will be amazing, the car will nag you much less.
“Nag” is what some people refer to as the alerts FSD gives drivers to stay attentive to the road.
Tesla requires all drivers to remain attentive to the road when using Autopilot or Full Self-Driving, and they are responsible for any mistakes made by the vehicle.
Musk has repeatedly promised that Full Self-Driving would eventually become unsupervised, but every time he shared a timeline, Tesla failed to deliver.
Electrek’s Take
We are talking about 6 weeks – meaning the end of September, which also happens to be the end of the third quarter.
Tesla will release the update just in time to justify recognizing some deferred revenue set aside from selling its Full Self-Driving package.
“Less nag” means nothing. 99% of the value of FSD is in unsupervised operation with responsibility on Tesla. That’s where the value lies, and we are still far from that.
Musk is now talking about FSD having “less nag” by the end of September. And then what? We should expect Tesla to go from that to taking responsibility for the system in customer cars by the end of the year? Please.
The best data available points to Tesla achieving about 400-500 miles between critical disengagements and the last time Tesla introduced a big parameter increase it resulted in roughly a 2x improvement.
At 800-1,000 miles between disengagement, Tesla would still be years away from unsupervised self-driving in customer vehicles.
Top comment by Johnny Nemo
I would love to see a correction to mistakes FSD has been making, reproducibly, at the same places that I drive every day on my commute, FOR THE LAST FIVE YEARS.
As far as "nagging" is concerned the "attention monitor" is flaky, and sometimes goes of as soon as I engage FSD. And yes, I am looking at the road, and I refuse to take of my sunglasses when driving with the sun in my eyes, especially as the FSD cameras also get blinded and panic demanding I take over immediately.
Look, I have FSD. I’m incentivized for Tesla to deliver here. I want to believe, but to do so, I can only rely on the latest in a long series of unfulfilled promises from a known liar versus actual data pointing the other way.
If Tesla sticks to this timeline, it would have been almost a year between v13 and v14, and based on previous improvements, I expect a 2 to 3x improvement in miles between disengagements.
At the bare minimum, Tesla would need a 50x improvement from there to deliver unsupervised self-driving in consumer vehicles and ideally more than 200x.
At the current pace of updates, we are years away from that.
FTC: We use income earning auto affiliate links. More.
Comments