Tesla full self-driving testers are reporting that FSD Beta 10 is a large improvement. They agree that in 2-4 weeks it will be ready for the start of wide opt-in release.
They are feeling far more confident and comfortable using FSD.
Gali of Hyperchange now thinks that robotaxi level could be reached in 6 months. Prior to version 10, Gali though that with version 9.2 it would take 5 years to reach robotaxi level. Gali thinks the current system is now safer than many human drivers.
It is not perfect. There are some testers who indicate some disappointment. Even those who are enthusiastic note a few disengagements and issues.
A wide release of FSD with full features that is able to handle almost all driving situations would enable more of the FSD revenue to be recognized. Currently, Tesla only recognizes about half of the FSD revenue. The rest of the revenue is deferred until the full features are available. Deferred revenue is payment that has been received for goods or services before they’ve been fully delivered.
Recognizing another 40% of the FSD revenue would be over $1 billion of revenue. This could get recognized in Q3 or Q4 of this year.
Brian Wang is a Futurist Thought Leader and a popular Science blogger with 1 million readers per month. His blog Nextbigfuture.com is ranked #1 Science News Blog. It covers many disruptive technology and trends including Space, Robotics, Artificial Intelligence, Medicine, Anti-aging Biotechnology, and Nanotechnology.
Known for identifying cutting edge technologies, he is currently a Co-Founder of a startup and fundraiser for high potential early-stage companies. He is the Head of Research for Allocations for deep technology investments and an Angel Investor at Space Angels.
A frequent speaker at corporations, he has been a TEDx speaker, a Singularity University speaker and guest at numerous interviews for radio and podcasts. He is open to public speaking and advising engagements.
No Tesla robotaxi is imminent.
There is a lot you can do with lvl4, look at waymo et al, but lvl5 isn't feasible with existing machine learning architectures.
The point is, it will be fully SD soon, but for legal reasons they will tell you that it isn't. So if you take your eyes off the road it will be fine, but it doesn't have to jump through regulatory hoops created by people who don't know what they have to regulate exactly since no one does.
Until it is filly SD for a stretch , so i can take my eyes of the road it is not only useless but actually dangerous since it promotes disengagement. When I am tired the drive assists help me fall asleep behind the wheel.
The Robotaxi problem is mostly a legal one. It pays for Tesla to say that their FSD system is "merely" a driver assistance until it wears off its novelty and judges are more liable to accept that it does work well enough that it can replace a driver. Until then, pushing towards robotaxi is likely to end up in a legal morass which hinders progress rather than helps.
Well, Elon does tend towards hyperbole every now and again. Then again, perhaps it is easier to be shocked at the performance when you know a lot about the difficulty of the problem. To a layperson, it drives. It takes an expert to notice all the little details which they know are really tough to solve that have been solved.
Clearly the early access fleet has not had their minds blown away as Elon suggested. But a tempered positive response is a good thing. This, plus the single stack merged street/highway code makes further progress potentially simpler to execute.
Yes those levels where a total distraction for average minds and journalists who thought you needed to go through all the levels as in a computer game. So their retarded reasoning was Tesla was far behind.
So you are also saying that there is no Tesla robotaxi in the horizon as this post suggests.
You may be right. That legal middle ground when the car is sometimes responsible and sometimes the driver, may be something that no automakers want to be in.
FSD Beta 10 seems like the deflection point. This level of performance from a pure vision NN is the starting point before the data flood from wide release in the wild, or Dojo 1 coming on line, or Hardware 4.
Despite all the other companies in the field, it seems quite likely that Tesla will scale into the Robotaxi space without any serious rival anywhere in sight.
The uselessness of the SAE ratings will be highlighted as Tesla FSD goes straight from “Level 2” to “Level 5”without ever pausing in between.
Don't confuse a name with the literal definition, recall Autopilot?
6 month or 5 years, I don't think that ever, other companies with a more wholesome approach to autonomy, are already rolling out robotaxi services. When hypothetically Tesla will supposedly reach robotaxi level of autonomy, other companies will have a solution as cheap as that as Tesla. A google search of "robotaxi news rollout roundup brought be Chinese startup Pony.ai, Waymo, hyundai, Ford, Lyft and Argo team, AutoX, and Apollo moon recent rollout news items in the first page.