Tesla FSD has enough compute and driving data for training so the limiting factor is how fast the system can be validated.
Only about 1/10,000 of distance driven is useful for training. This means of the 1 billion miles per month that is gathered by Tesla from over 1 million cars using FSD, there is about 100,000 miles of interesting data for training.
Validation means testing a new version after it’s been trained to verify that it is good enough to give to customers.
Accurate analysis.
Only about 1/10,000 of distance driven is useful for training.
The limiting factor for FSD progress was AI training compute, but now it is validation, as the interventions are so rare. https://t.co/SL0FEZ6zqk
— Elon Musk (@elonmusk) May 7, 2024
Tesla has the data to determine how much training is needed to correct a known problem. Tesla is using the next two or three versions of FSD and knows how good they will be. Tesla also has good projections of what FSD capabilities will be as the compute, training, models and architecture is improved.
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.
I think they’re depending on the demand curve for taxi services to be such that if you lower the prices significantly (which can be done by removing the need for paying a human driver) that the demand will be orders of magnitude higher. I can see this being the case if the cost was low enough that it saved money compared to owning a car.
I’m puzzled. Taxis, Ubers are a rather small part of the vehicles on the road. How does Tesla plan to make a profit from that market segment, particularly when plowing in huge amounts of development money?
Throw in the reluctance of the public to buy electric vehicles and the future of Tesla looks pretty sketchy. Not providing the necessary charging infrastructure puts another wet blanket on Tesla’s financial future.
If the limiting factor is validation, what is stopping Tesla from 10x the validation team? Hiring, say, a thousand drivers?
Why hire drivers when Tesla car owners work for free as there drives are what is used for training FSD.
Ugh. Has SolvingTheMoronProblem stopped calling everyone morons and doing stupid voices yet?