Error in title: He did not say level 5. Reported, since can't delete or edit title.
Last edited:
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Elon in 2019 saying for sure, level 4 by end of 2020
- Thread starter DanCar
- Start date
Error in title: He did not say level 5. Reported, since can't delete or edit title.
I thought it was pretty common knowledge around here that you have to add at least a few years to any future prediction he makes.
EVNow
Well-Known Member
When it comes to FSD - not just Musk, you have to add a few years to everyone's predictions !
ps : Tech industry has a lot of examples where people highly overestimated progress possible in the short term - but underestimated the possibilities in the long term.
What is the purpose of this thread? It has nothing to do with AP.
Mods, I'm BEGGING YOU can we get ONE thread for people to complain about Elon missing promised dates? EVERY thread now has one or two people clogging them up with the same B.S. over and over. I get it, it is taking longer then he said, I don't need every thread to remind me over and over. It has completely diluted any useful discussion on this forum.
Mods, I'm BEGGING YOU can we get ONE thread for people to complain about Elon missing promised dates? EVERY thread now has one or two people clogging them up with the same B.S. over and over. I get it, it is taking longer then he said, I don't need every thread to remind me over and over. It has completely diluted any useful discussion on this forum.
Tam
Well-Known Member
I think it's about salesmanship rather than scientific presentation.
If it was a technical presentation, it would cover why the present Tesla system still collides with obstacles from slow speed in smart summon to high-speed freeways.
After presenting the cause, it would cover a demonstration of current technology that would solve the collision problems.
It's not a good explanation to say the AI, Neural Net, HW4 would take care of it. Because that means it's not even getting a passing grade in hundreds/thousands of real-life testing of a parking lot or parking police cars on highway just yet.
It's hard to predict something that has not passed even a number of preliminary tests.
MarsOrBust
Member
He did not qualify the ‘sleeping in your car’, so it must assume you are travelling to any location, hence L5 by implication
You could argue robotaxis are only L4, as they can be limited to & typically operate in cities
Regardless, as my video post here shows, he has been making autonomous driving promises since 2014:
https://teslamotorsclub.com/tmc/threads/a-video-summary-of-elon’s-broken-deadlines-predictions-for-fsd.224038/
I'm still waiting for my AP1 car to pick me up anywhere on my property. This is taking a long time. Hope it happens soon. I'm starting to get tired of waiting.
Tam
Well-Known Member
Since I am not an expert so it's easy for me to rely on what Tesla predicts. But I notice that the timeline keeps on slipping that's when I realize that:
Yes, the rocket was first done in 1926 so it's easy to predict humans can colonize Mars. But there are more things to deal with than the first 1926 rocket in order to get to Mars. And it's been almost a century dream.
The facial recognition puzzle has been worked on since the 1960s. After about 60 years, many cities are banning it. The AI in facial recognition works great and very amazing in general but that is still not good enough to prevent false arrests and racial bias.
It's just like since AI can solve the problem of reading the alphabets, it's easy to predict that it can solve the Autonomous Vehicles but I think once it starts to work on solving the problem of Autonomous Vehicles, that's when the prediction would take longer, then longer.
Hey! There's a parked police patrol car over there on the freeway: how will you train AI, Neural Net, HW4 not to hit it this time?
Tam
Well-Known Member
It's interesting that Theranos CEO was photoshopped together with Tesla CEO:
In the old days, to test your blood glucose level, a full tube of blood must be drawn. Now, you only need a pinprick for a drop of blood.
Theranos wanted to expand that success to other tests as well.
It's not like the blood test technology is non-existing. Laboratory technology works. What Theranos proposed was to make it work better by shrinking the volume requirement of the specimens just like the way it happened with testing your blood sugar level.
Ok! It's true that it doesn't work now for other tests and the timeline has passed but that doesn't mean that the technology won't ever evolve and be successful in the future.
Back to Autonomous Vehicles. Technology has been around since the DARPA Grand Challenge as far back as 2005. Tesla proposes that instead of being burden with all the crutches of LIDAR... it can simplify the task to vision, camera base, it's like superhuman eyesight. Like Theranos, people might become impatient as the timeline has passed but that doesn't mean it won't ever be successful in the future.
Same with going to Mars. The timeline might have passed but that doesn't mean the humans won't ever be able to go to Mars.
So? Missing timeline means scam now?
In the old days, to test your blood glucose level, a full tube of blood must be drawn. Now, you only need a pinprick for a drop of blood.
Theranos wanted to expand that success to other tests as well.
It's not like the blood test technology is non-existing. Laboratory technology works. What Theranos proposed was to make it work better by shrinking the volume requirement of the specimens just like the way it happened with testing your blood sugar level.
Ok! It's true that it doesn't work now for other tests and the timeline has passed but that doesn't mean that the technology won't ever evolve and be successful in the future.
Back to Autonomous Vehicles. Technology has been around since the DARPA Grand Challenge as far back as 2005. Tesla proposes that instead of being burden with all the crutches of LIDAR... it can simplify the task to vision, camera base, it's like superhuman eyesight. Like Theranos, people might become impatient as the timeline has passed but that doesn't mean it won't ever be successful in the future.
Same with going to Mars. The timeline might have passed but that doesn't mean the humans won't ever be able to go to Mars.
So? Missing timeline means scam now?
That the promised dates are regularly pushed back & the definition is walked back (L4 is often defined as L5), suggests that it might not be able to be done, ever
As per the YouTube video, the side B pillar cameras could struggle with forward cross traffic because they are set too far back (further than the typical position of a driver’s eyes). Additionally, I’ve heard it suggested that the rear camera fails to see rear cross traffic, when backing out from between 2 obstacles (eg: parked cars)
MarsOrBust
Member
I don’t think missing a timeline means scam but constantly moving the goal post when you keep chirping it’s just around the corner at the very least smells fishy.
rxlawdude
Active Member
I actually use the Theranos case in my Bioethics course to students going into pharma or device manufacturing industries. The problem with the Theranos device was obvious to anyone with a modicum of understanding clinical lab science. You simply cannot minaturize an entire lab with 200+ test types in the confines of a box the size of a laser printer. Not now. Not in the wettest dream of the future.
In other words, Holmes's pitch was never attainable and she understood that. (They even cheated potential investors/customers by taking their blood, putting it into the machine, and then taking the group to lunch. Meanwhile, the people from the real but hidden lab downstairs would run up, grab the samples, run them in the basement and then make it appear the analyses were done on the machine. That's clear fraud.
Elon's statements are aspirational (and optimistic!), but he clearly has a goal and the technical chops to make FSD a reality.
Apples and orange comparison between these two companies and their leaders.
Similar threads
