Last weekend Tesla FSD 12.4.1 was expected to go to wide release but it did not. This weekend if there is no safety regression there should be a push of FSD 12.4.2.
It is reported that 12.4.2 is much smoother. The challenge is keeping safety high, while increasing ride comfort. The 12.4.2 build moved into internal testing today.
There is still progress and fairly rapid releases of FSD but we are still waiting for the breakthrough release.
I think the major breakthrough levels will be one that gets over 1500 miles of good driving between interventions. This would be more than one month of driving for one regular individual.
The next major breakthrough level would be getting over 15,000 miles of good driving between interventions. This would be more than one year of driving for one regular individual.
I think getting over 150,000 miles of good driving between interventions would be beyond human safety since most interventions would involve preventing life threatening accidents. the 15k-150k level would then have high confidence robotaxi with safety driver or remote driver monitoring.
12.4.2 is much smoother. Challenge is keeping safety high, while increasing ride comfort.
Build moves to internal testing tomorrow.
— Elon Musk (@elonmusk) June 26, 2024
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.
Any version from here out is probably viable with geofencing.
Tell the router to avoid unprotected left turns during busy hours and a handful of predictable edge cases. I avoid unprotected left turns and construction areas when I can. It is safe driving behavior to pick a safe route.
The emergent edge cases it struggles with seem rare enough for commercial operations.
Since we have no way to figure out which one of the 15k miles you will have to take over, you will have to stay alert at all times, which is identical to driving yourself. Robotaxi is useful ONLY if you can do something else than staying alert.
I agree, as long as I must pay attention all the time, I also don’t see it as very useful tech. If I can turn on autopilot and sit in the back seat, have a beer/ surf the web then we can say we have proper SD tech.
Hopefully we will get there around 2027-2030
True at 15k. Maybe not 150k. At some point the numbers represent that qualitative shift to unsupervised where the system is safer without any supervision than a good human driver and any further delays in regulatory approval of driverless operation unambiguously are costing lives.
No. Statistical analysis are tricky: having an intervention frequency of 1 intervention every 150k miles means that every 150k drivers driving for a mile there is a statistical certainty of one issue in the first mile.
Being a passenger in a vehicleto make sure the driver doesn’t fall asleep is the same thing as driving a vehicle. /s
FSD 12.4 is not based on FSD 12.3 and has way more data. This means, it takes more time or better instruments to get rid of bad data. Very looking forward to a polished FSD 12.4 version.