This video was sponsored by Brilliant. Sign up at brilliant.org/... for a free 30-day trial + 20% off the annual premium membership! Consider supporting the channel / aidrivr
Hey, that was my Silver Model 3 at 3:30...yes RIP, was totalled when rear-ended by the the guy behind us who must have forgotten his Nissan van didn't have any driver assist features.
8:24 to 8:34 looks more like collision avoidance than "aggression" to me. While borderline blowing through a yellow doing 46 in a 35, a car pulls out in front and it's like "uh-oh, speed differential." Then the other car speeds up and you see the path planner go "oh, okay."
I think it’s also worth mentioning the recent revelation that Tesla’s been manually tracking and updating areas where influencers drive to skew what audiences are seeing
One of the side effects of using ML based algorithms is that seemingly insignificant variations in conditions (of which there are plenty while driving) might result in different behavior. This would look like inconsistency to a driver. It's why it's so important to get enough randomized training data to erase these conditions. In other words, I think this inconsistency you're noticing is due to their stated lack of training on normal driving.
More randomized normal driving data is just going to shift inconsistency to the edge cases. ML training has never shown to work on such large models. Large in scope of the problem. ML models are always specific and limited in scope. No amount of data or training will fix the inconsistency/hallucinations because they are inherent features of these fast but non-robust models.
You should make a video talking about Tesla workers putting extra work into areas that FSD influencers like yourself live, to make FSD look better than its general performance actually is.
What a great summary of how FSD is working today. I have 12.3.6 on my daily driver, and 12.4.3 on my other Tesla. AI DRIVR sums up my experience much better than I ever could. AI DRIVR's video overlay of FSD graphics with view out the windshield is by far my favorite format to watch FSD, and his commentary is absolutely top notch.
ChatGPT hallucinations are not fixed by more data but by stacking different models. Like chain of thought, RAG, even calling up python codes. Hallucination is a feature of these algorithms not a bug to iron out. It’s inherent to the math functions. More advanced models can fix it, but those models are slow. FSD can’t sit and wait for answers.
@@AudiTTQuattro2003 computation resources to handle a more complex array of NN models like ChatGPT. When FSD comes up to a Chuck Cook’s left turn, it needs to switch to a NN built for that type of turn only. You can’t train those turns on a large NN not expect it to mess up other types of turns. So the onboard system needs to be powerful enough to run many models in parallel. At least 10x of the current system.
I've experienced extremely hard braking on v12.3.6 when approaching an intersection and the traffic signal changes to yellow. Extremely hard braking. I was lucky no one was behind me because I would have experienced a rear-end collision and/or a pissed off driver behind me. I've seen others report this as well. From a technical perspective, FSD may be correct in stopping however hard is necessary when approaching an intersection with a yellow light, but it should also take into account the g-force that would result if it decides to come to a stop instead of continuing on thru the intersection. I hope v12.5 fixes this issue.
Thanks for another great vid. I think sometimes you are giving it undue credit for thinking. For example, at 4:41, it's simply not turning left because the light is green and it doesn't have right of way. I also don't think it understands about in-n-out waiters taking orders. It's just hesitant to move in case the pedestrian is going to walk into the road.
Most if the driving people do is local. When we drive in unfamiliar areas we make loads of mistakes, driving in the wrong lane for a turn off, missing turn offs, pissing off locals... FSD has to drive like this all the time. I see loads of testers say things like, its slowing down too aggressively, its approaching the stop sign too slowly, its stopping too far back, its not pulling into the lane quickly enough to.make the next turn, etc. This is probably all down to the tester knowing the local traffic behaviour. It will be a while before Tesla start including local data for their drives. Until then, drivers will be critical. Passengers probably won't care though, so we need to jump to the part where supervision is not needed and Passengers don't notice if the car misses a turn or other triviality.
Has Tesla said they will eventually add local data? Seems like a no brainer at this point? Why haven’t they done it yet? Explain like I’m 12 yrs old pls
@@SoccerScienceLabprobably because every car would have to get its own set of training data module. Impractical at this point in the game. Maybe once the wide release model is good enough, some of the DoJo compute power can be utilized for tailored applications.
As long as it doesn't crash without the driver and arrives at the destination. It's good enough. But that is not the case right now, so it's just a driver assist. Comfort is a nice to have, it's not a concern.
Passengers very much notice if other drivers are honking at or crashing into the car, so drastically uncommon acceleration is a problem. This is why NHTSA mandating a full stop at stop signs is probably the most dangerous thing they ever did.
@@SoccerScienceLab Specialized training for local areas is a last step. We're still in the general training phase, where the overall behavior is changing drastically. Trying to add finishing touches on an unstable foundation would be counter-productive - like localizing the language for each country on a website that is still in development, with the text constantly changing. Much less work to wait until it's stable.
I’m on 12.3.6 and the other day it completely blew a normal left hand turn that other versions have never had trouble with. The car went WAY wide and was headed off of the road so I had to take control. Pretty freaky.
1:40 this is one of the most annoying thing my car does nowadays. I hate cutting in line but it often waits until last minute to do the lane changes even though there was good times to make the change earlier. 2:20 the unusual lane changes also bothers me too. my 12.4.3 does both of these still
I've now been driving 12.4.3 a lot and you sum it up well. It's mostly OK but there are times that make you scratch your head and say "WTF... REALLY?" For me overall, 12.4.3 is worse than 12.3.6. 12.4.3 hesitates way too much turning from one road onto another and the auto speed control is TERRIBLE now. Most of the time it'll go way under the speed limit and you have to remind it to go faster with the accelerator and THEN it'll maintain the speed limit. Then in other sections, the speed limit goes from 60 to 45 and it'll just keep driving 65. Looking forward to 12.5 as this is 1 step forward, 2 steps back.
@@AIDRIVRyes. This is definitely a thing as we non-influencers are very aware of. We do not see this level of full self driving. You can see this in action with Chuck Cook’s videos showing Tesla testing the unprotected left in his area.
I expect we just got lucky - it almost certainly has no understanding that a person standing next to a car at a drive-through means they're taking an order and you should wait until they move on. But it is interesting. I wonder what the rationale was; maybe just too close to the car for it to feel confident moving? 🤷♂
For the first tieme in 4 yrs of FSD 12.3.6 read route signs on a night drive as speed limit signs . slowing me down to 30mph n a 60 mph zone. Did this multiple times forcing me to disengage FSD
Love your videos! Great that you’re highlighting the good and the bad including the multiple WTF scenarios, instead of just the maneuvers it handles well. I have 12.4.3 myself and have encountered many of the issues you indicated. Nicely done!
Earlier 12s were way better. I subscribed last month and it has gotten so bad I had to stop using it. Would just sway back and forth like it’s drunk on the road and wouldn’t pick a lane and constantly tried to get in the right lane that was ending.
Someone internal at Tesla mentioned Tesla employees have been "crafting" FSD for Musk and "celebrities". So I am highly skeptical of some of these videos. There is a guy driving in Chicago with FSD and it is a complete trash. The car does almost nothing by itself.
There is a big enough body of evidence to allay any fear of some sort of "fix" going on. There is a free month for every Tesla driver in America. If it really had problems, there would be a mountain of evidence...
One problem I’ve noticed is that the cars don’t seem to recognize when they are out of calibration. Having driven at 4 different model Xs in the past month, the variation in state of calibration has severe impacts on the FSD performance. I realize I can remedy this, but robocars would not.
Which brings the point, who is responsible if the calibration is off in a robotaxi? The owner? Tesla? If it causes an accident, who pays for damages? Why can't the system self calibrate?
With the knowledge that Tesla especially trains FSD for influencers and people showing of FSD this showcase is extremely bias and is probably not representative of the actual performance of the model. You would need to drive in a complete new area everytime so that you have actual performance representation.
The inconsistency seems logical. The system is testing in scenarios where sometimes users disconnect and sometimes they 10:31 don’t. Our reactions and/or its own results on navigating feed the net.
It drives ok but can not drive properly in heavy traffic. Most importantly still doesn’t know blinking green means protected left turn, also makes turns on straight arrow which is illegal. It can not be more basic than that.
But we now know the Tesla specifically targets influencers with better self driving ai training to give them a better experience than the average owner.
Great summary of these last few weeks filled with MONTHS worth of FSD updates. Truly looking forward to 12.5, Robotaxi roadmap and FSD release in Europe. I suggest you could also do a robotaxi-ready timeline prediction based on your tons and tons of experience with the software. Anyways, always a pleasure to learn from the master!
The fact that it has completely ignored solid street lines and dropped my tires off of a mountain means that going purely inference-only is a horrible decision. There needs to be some of those 50k lines back in to consistently prevent that kind of behavior.
Machine learning is just a black box. input goes in and output goes out, but we have no clue why it chose the output it did, and that's also why it's so inconsistent. Nothing smart is happening inside that silicon, and who knows why it does what it does.
Your section about how it used to work and how it works now "300,000 lines of code, blah blah" is pure Elon bullshit. You can ask Andrej Karpathy how it actually worked. Anyway, this inconsistency is exactly why Tesla is so far from a true robotaxi.
Exactly. If the Crowdstrike debacle teaches us anything, it is that a simple human error could render a new update or version dangerous, unless it is completely vetted in a controlled environment. The AI still doesn't know what it doesn't know from one version to the next.
There is a "Do Not Block" sign and clear pavement markings at a fuel station I pass leaving my neighborhood. V12.4.3 was ready to totally ignore the signs and markings two different times, forcing me to hit the brake and disengage.
There are some scenarios and intersections where even school busses cross double yellows particularly when there are two sets of double yellows with usable pavement inbetween that is needed to keep traffic flowing.
I think telsa needs to give testers like you the ability to save both good and bad data, then give it a few labels yourself (where you group things together, like being in the intersection with a green or that intersection that failed first go but succeeded 2nd go) you group the similar situations and label some as bad and some as good that way they have the good data to counteract the bad data.
Exactly. It may work flawlessly by then, or we might come to find out it can't because of too much bounce between updates from new "weights" in the algorithm cause undesirable consequences elsewhere.
I think Tesla is honing these releases to excel in areas traveled by reviewers. I can’t go half a mile from my house without an intervention. Indecision, accelerating through dips, aggressive acceleration, frequent lane changes (still there for me), inability to make decisions when no other cars around. It’s not good. Better. But not good.
At least when FSD does kill some beta tester or innocent bystander, the lawyers will have mountains of evidence to win civil suits. The old "the beta tester is responsible" argument will eventually wear thin.
This is probably the biggest issue FSD still has. I love it, but slowing down to 40 on a 75+ mph road because there was a sign for "Route 404" is just something that can't happen. That and stopping for the wrong red light, e.g. one on an overpass running parallel to the road the car is actually on, are the two things making me really nervous for the RoboTaxi. Both of these things have happened to me in my Model 3! It's only that last
Just moved out of the neighborhood you always drive around in Berkeley / Kensington. So nice to have normal streets now lol those ones are ridiculous even for normal drivers
My Y running 12.4.3 just learned how to turn into the driveway and line up with the garage but as of yet, won't go in by itself. Maybe I can train it to pull into the garage.
I wouldn't want ChatGPT driving for me. Too many examples of it producing lists of words that end with an n, but actually don't. And then fixing the list incorrectly when that's pointed out.
This video seems really interesting. Currently I’m looking at buying one of these. It doesn’t seem really like it’s worth $12,000 extra. Maybe the future. It is but right now it seems like you are clearly buying for the future.
FSD works well in city driving. Probably because all the testers drive these areas. They need to spend some time on rural roads where it consistently fails to handle basic traffic scenarios. An unprotected left turn on a 25-35 MPH road is not the same as a 55 MPH road. More than anything else, they need to work on consistency. I should be able to know how the car is going to react so I am prepared to react. 12.4, for me, is a big step backward from 12.3 in most ways.
For seven years now the improvements in FSD have been huge, awesome, fantastic. Still we are not done. There is not one driverless Tesla to date. 6 years ago I bought a model S with FSD (in Europe), I sold it before I could ever see FSD in action. I no longer drive Tesla (I now drive Xpeng, its FSD is free because it is just as powerful as Teslas and they do not consider that complete). I do like to watch your videos though and keep up to date on the ever lasting experiment. Being frustrated about the past 6 years though the powerful words you use on the progress … hurt. Suggestions for better words: tiny improvements, some improvements but no where near the end product, baby steps as usual. In other words, a little bit less positive would be fair, I think. Unless … you get payed by Elon … in that case, continue with the big words.
It isn't necessarily being payed, but the fact that theTesla influencers might loose access to God, oops I mean Elon, which would destroy their ability to produce "breaking" content.
As good as it is, would you want to insure a car with FSD and no human to supposedly take over in case it screws up? It has to be better than a sober, attentive human in every way, or it isn't ready.
Not sure if it was a mic change or something else, but I just noticed your commentary style changed a lot the last couple of videos. It used to be a lot more relaxed, easy to listen to. Now it's way more pronounced but at the same time sounds more "yelling"-like and is tougher on the ears, at least for me.
I'm still on FSD 12.3.6 and I have been trying to use it for four days. In the neighborhood but yesterday I got out of my comfort zone tried it on a main road kinda like highway like. So when will I get 12.4 and 12.5?
The craziest part is 12.5 will probably release this year, and that's going to be another big shift as it moves FSD's highway stack from hand-coded to AI.
Anyone who has FSD here it would be awesome if you could report your data to FSD community tracker! It is the only way we can continue to follow the improvement trend as it gets better and better!
You have to have your hands on the steering wheel at all times when you are driving.If you don’t have the car will not drip which to me is necessary.There have been a lot of deaths all over the world because of this and dangerous incident.Not long ago two people where I live died in a crash in a self driving Tesla which drove into a three. Another incident from a Tesla where it suddenly would change lane without the consent of the driver and he had turn the wheel hard to avoid a crash .The first step was the autopilot which now can keep the distance to other cars while driving . Automations can be dangerous as quoted.
Gotta stop you there, just two minutes in. To say, great intro, beautifully done, tho' I would expect no less from you. Can't wait for the rest, ok onwards...!
Hey, that was my Silver Model 3 at 3:30...yes RIP, was totalled when rear-ended by the the guy behind us who must have forgotten his Nissan van didn't have any driver assist features.
nice!
rip.
No better a way to ruin your day. :(
Will you buy a new Highland?
@@david78bemaybe, have a Cybertruck to drown my sorrows for now😅
The way it handles the do not block intersection sign and the drive thru is crazy
The drive thru really blew my mind. I also thought that we would have to wait months if not years for this level of intelligence...
i feel like the drive through wasn't FSD understanding what was happening, but rather refusing to move forward because a pedestrian was so close.
@@whetlarper Wait! You can't say that! This is pure Elon 4d Chess driving! These cars are magic! Robotaxi here we come!!!
It just did not want to stand in a 4 way intersection, nothing special.
@@jasonk125 yeah it's so bad bro it is only driving on its own slightly worse than the average college student you totally owned all the elon bros
8:24 to 8:34 looks more like collision avoidance than "aggression" to me. While borderline blowing through a yellow doing 46 in a 35, a car pulls out in front and it's like "uh-oh, speed differential." Then the other car speeds up and you see the path planner go "oh, okay."
Agreed. If it were me I either would've gotten over or slowed significantly.
I think it’s also worth mentioning the recent revelation that Tesla’s been manually tracking and updating areas where influencers drive to skew what audiences are seeing
It’s like shipping a “golden sample” of a product to a reviewer or, rather, shipping a particularly suped-up sample.
Any particular Drive-Fluencers come to mind. Would be curious to watch these
One of the side effects of using ML based algorithms is that seemingly insignificant variations in conditions (of which there are plenty while driving) might result in different behavior. This would look like inconsistency to a driver. It's why it's so important to get enough randomized training data to erase these conditions.
In other words, I think this inconsistency you're noticing is due to their stated lack of training on normal driving.
More randomized normal driving data is just going to shift inconsistency to the edge cases. ML training has never shown to work on such large models. Large in scope of the problem. ML models are always specific and limited in scope. No amount of data or training will fix the inconsistency/hallucinations because they are inherent features of these fast but non-robust models.
You should make a video talking about Tesla workers putting extra work into areas that FSD influencers like yourself live, to make FSD look better than its general performance actually is.
Missed your vids, my fave and only source I really need to feel up to date.
BABE WAKE UP AI DRIVR JUST DROPPED
But I have already finished watching, honey... When 12.5???
Right?! I have been waiting for the latest drop 😂
Go back to bed. Another fail. 😂
What a great summary of how FSD is working today. I have 12.3.6 on my daily driver, and 12.4.3 on my other Tesla. AI DRIVR sums up my experience much better than I ever could.
AI DRIVR's video overlay of FSD graphics with view out the windshield is by far my favorite format to watch FSD, and his commentary is absolutely top notch.
ChatGPT hallucinations are not fixed by more data but by stacking different models. Like chain of thought, RAG, even calling up python codes. Hallucination is a feature of these algorithms not a bug to iron out. It’s inherent to the math functions.
More advanced models can fix it, but those models are slow. FSD can’t sit and wait for answers.
In some way, computational resources will always be the limiting fact. Thus the question becomes, at what hardware level will FSD become acceptable?
@@AudiTTQuattro2003 computation resources to handle a more complex array of NN models like ChatGPT. When FSD comes up to a Chuck Cook’s left turn, it needs to switch to a NN built for that type of turn only. You can’t train those turns on a large NN not expect it to mess up other types of turns. So the onboard system needs to be powerful enough to run many models in parallel. At least 10x of the current system.
Never been so early. Was waiting for this video for daaays 🔥
Yes bro same to me
I've experienced extremely hard braking on v12.3.6 when approaching an intersection and the traffic signal changes to yellow. Extremely hard braking. I was lucky no one was behind me because I would have experienced a rear-end collision and/or a pissed off driver behind me. I've seen others report this as well. From a technical perspective, FSD may be correct in stopping however hard is necessary when approaching an intersection with a yellow light, but it should also take into account the g-force that would result if it decides to come to a stop instead of continuing on thru the intersection. I hope v12.5 fixes this issue.
I like that you decided to talk about each version in the video, rather than skip the intermittent versions because a new version came out
Thanks for another great vid. I think sometimes you are giving it undue credit for thinking. For example, at 4:41, it's simply not turning left because the light is green and it doesn't have right of way. I also don't think it understands about in-n-out waiters taking orders. It's just hesitant to move in case the pedestrian is going to walk into the road.
Do you think Tesla has specially trained their self-driving model to the area you are testing this in?
Most if the driving people do is local. When we drive in unfamiliar areas we make loads of mistakes, driving in the wrong lane for a turn off, missing turn offs, pissing off locals... FSD has to drive like this all the time. I see loads of testers say things like, its slowing down too aggressively, its approaching the stop sign too slowly, its stopping too far back, its not pulling into the lane quickly enough to.make the next turn, etc. This is probably all down to the tester knowing the local traffic behaviour. It will be a while before Tesla start including local data for their drives. Until then, drivers will be critical. Passengers probably won't care though, so we need to jump to the part where supervision is not needed and Passengers don't notice if the car misses a turn or other triviality.
Has Tesla said they will eventually add local data? Seems like a no brainer at this point? Why haven’t they done it yet? Explain like I’m 12 yrs old pls
@@SoccerScienceLabprobably because every car would have to get its own set of training data module. Impractical at this point in the game. Maybe once the wide release model is good enough, some of the DoJo compute power can be utilized for tailored applications.
As long as it doesn't crash without the driver and arrives at the destination. It's good enough.
But that is not the case right now, so it's just a driver assist. Comfort is a nice to have, it's not a concern.
Passengers very much notice if other drivers are honking at or crashing into the car, so drastically uncommon acceleration is a problem. This is why NHTSA mandating a full stop at stop signs is probably the most dangerous thing they ever did.
@@SoccerScienceLab Specialized training for local areas is a last step. We're still in the general training phase, where the overall behavior is changing drastically. Trying to add finishing touches on an unstable foundation would be counter-productive - like localizing the language for each country on a website that is still in development, with the text constantly changing. Much less work to wait until it's stable.
Whats your take on Tesla specifically developing FSD for influencers? Using footage from creators videos to enhance your specific driving experience?
I’m on 12.3.6 and the other day it completely blew a normal left hand turn that other versions have never had trouble with. The car went WAY wide and was headed off of the road so I had to take control. Pretty freaky.
1:40 this is one of the most annoying thing my car does nowadays. I hate cutting in line but it often waits until last minute to do the lane changes even though there was good times to make the change earlier.
2:20 the unusual lane changes also bothers me too.
my 12.4.3 does both of these still
I've now been driving 12.4.3 a lot and you sum it up well. It's mostly OK but there are times that make you scratch your head and say "WTF... REALLY?" For me overall, 12.4.3 is worse than 12.3.6. 12.4.3 hesitates way too much turning from one road onto another and the auto speed control is TERRIBLE now. Most of the time it'll go way under the speed limit and you have to remind it to go faster with the accelerator and THEN it'll maintain the speed limit. Then in other sections, the speed limit goes from 60 to 45 and it'll just keep driving 65. Looking forward to 12.5 as this is 1 step forward, 2 steps back.
Have you heard the claims that Tesla specially addresses the areas around reviewers to give them extra performance. What do you think about that?
want me to make a video about it?
@@AIDRIVR ommm, Yes.
@@AIDRIVR It's extremely ridiculous, I think making a video about that would be worthless...
@AIDRIVR Yeah, would be cool to hear your opinion on that
@@AIDRIVRyes. This is definitely a thing as we non-influencers are very aware of. We do not see this level of full self driving. You can see this in action with Chuck Cook’s videos showing Tesla testing the unprotected left in his area.
Woah the behavior at the drive through was impressive
I expect we just got lucky - it almost certainly has no understanding that a person standing next to a car at a drive-through means they're taking an order and you should wait until they move on. But it is interesting. I wonder what the rationale was; maybe just too close to the car for it to feel confident moving? 🤷♂
@@catbert7 You're probably right. Maybe coincident or something.
Always enjoy your posts. Glad to see you working on it!
INCREDIBLE summary! Thank you.
For the first tieme in 4 yrs of FSD 12.3.6 read route signs on a night drive as speed limit signs . slowing me down to 30mph n a 60 mph zone. Did this multiple times forcing me to disengage FSD
No mention of potential VIP treatment for FSD?
Love your videos! Great that you’re highlighting the good and the bad including the multiple WTF scenarios, instead of just the maneuvers it handles well. I have 12.4.3 myself and have encountered many of the issues you indicated. Nicely done!
Earlier 12s were way better. I subscribed last month and it has gotten so bad I had to stop using it. Would just sway back and forth like it’s drunk on the road and wouldn’t pick a lane and constantly tried to get in the right lane that was ending.
Someone internal at Tesla mentioned Tesla employees have been "crafting" FSD for Musk and "celebrities". So I am highly skeptical of some of these videos. There is a guy driving in Chicago with FSD and it is a complete trash. The car does almost nothing by itself.
There is a big enough body of evidence to allay any fear of some sort of "fix" going on. There is a free month for every Tesla driver in America. If it really had problems, there would be a mountain of evidence...
There is no question that a large amount of the training data is from California Nevada Arizona and Texas.
I've been waiting for this for so long
Best Tesla channel by far!!! Keep up the great work!
Excellent video. Your visualizations, editing and commentary are the best! 👍 Thanks.
FSD handling drive-through ordering at In-n-out is really surprising.
Cool now it will get tickets for you!
One problem I’ve noticed is that the cars don’t seem to recognize when they are out of calibration. Having driven at 4 different model Xs in the past month, the variation in state of calibration has severe impacts on the FSD performance. I realize I can remedy this, but robocars would not.
Which brings the point, who is responsible if the calibration is off in a robotaxi? The owner? Tesla? If it causes an accident, who pays for damages? Why can't the system self calibrate?
With the knowledge that Tesla especially trains FSD for influencers and people showing of FSD this showcase is extremely bias and is probably not representative of the actual performance of the model.
You would need to drive in a complete new area everytime so that you have actual performance representation.
The inconsistency seems logical. The system is testing in scenarios where sometimes users disconnect and sometimes they 10:31 don’t. Our reactions and/or its own results on navigating feed the net.
It drives ok but can not drive properly in heavy traffic. Most importantly still doesn’t know blinking green means protected left turn, also makes turns on straight arrow which is illegal. It can not be more basic than that.
But we now know the Tesla specifically targets influencers with better self driving ai training to give them a better experience than the average owner.
Great summary of these last few weeks filled with MONTHS worth of FSD updates. Truly looking forward to 12.5, Robotaxi roadmap and FSD release in Europe.
I suggest you could also do a robotaxi-ready timeline prediction based on your tons and tons of experience with the software. Anyways, always a pleasure to learn from the master!
Thank You for All that you are doing for our Planet Earth.... Peace.. Shalom.. Salam.. Namaste .. 🙏🏻 😊 ✌ ☮ ❤
The fact that it has completely ignored solid street lines and dropped my tires off of a mountain means that going purely inference-only is a horrible decision. There needs to be some of those 50k lines back in to consistently prevent that kind of behavior.
Machine learning is just a black box.
input goes in and output goes out, but we have no clue why it chose the output it did, and that's also why it's so inconsistent.
Nothing smart is happening inside that silicon, and who knows why it does what it does.
maby some of that inconsitency is comming from Tesla focosing on routs that youtuber like to drive
lmfao 4:26 u drive by my house and yes that intersection is so funny to watch FSD just go ham on
Your section about how it used to work and how it works now "300,000 lines of code, blah blah" is pure Elon bullshit. You can ask Andrej Karpathy how it actually worked. Anyway, this inconsistency is exactly why Tesla is so far from a true robotaxi.
Exactly. If the Crowdstrike debacle teaches us anything, it is that a simple human error could render a new update or version dangerous, unless it is completely vetted in a controlled environment. The AI still doesn't know what it doesn't know from one version to the next.
There is a "Do Not Block" sign and clear pavement markings at a fuel station I pass leaving my neighborhood. V12.4.3 was ready to totally ignore the signs and markings two different times, forcing me to hit the brake and disengage.
I think 12.4 hallucinates too much. just like gpt in its early days
please test more at night and in the rain
...or driving straight into sunlight just above the horizon.
There are some scenarios and intersections where even school busses cross double yellows particularly when there are two sets of double yellows with usable pavement inbetween that is needed to keep traffic flowing.
I think telsa needs to give testers like you the ability to save both good and bad data, then give it a few labels yourself (where you group things together, like being in the intersection with a green or that intersection that failed first go but succeeded 2nd go) you group the similar situations and label some as bad and some as good that way they have the good data to counteract the bad data.
The new Crazy Taxi looks really good!
George Hotz over at Comma felt this was the way to go all along.
theese models NEED some muscle-memory
Exactly. There seems to be no relationship between versions about absolute "best practices". Tweaks here, breaks something else there.
12.5 video, when?
"inconsistency" is and has been a fundamental failure. It's why I largely avoid the FSD I paid for. Without FSD is less stress than with.
Some real drivers are even worse
Very depending on where you live right now.
Still 5 years away from eyes off and brain off.
Exactly. It may work flawlessly by then, or we might come to find out it can't because of too much bounce between updates from new "weights" in the algorithm cause undesirable consequences elsewhere.
Always back for more. Thanks! 12.5.x incoming..
The version number should be 0.xx until it is fsd.
Trained too much on Worst driving, Tesla: Becomes agressive
That FCW at 6:55 man how I hate bs FCW and how it destroys my insurance
I think Tesla is honing these releases to excel in areas traveled by reviewers. I can’t go half a mile from my house without an intervention. Indecision, accelerating through dips, aggressive acceleration, frequent lane changes (still there for me), inability to make decisions when no other cars around. It’s not good. Better. But not good.
At least when FSD does kill some beta tester or innocent bystander, the lawyers will have mountains of evidence to win civil suits. The old "the beta tester is responsible" argument will eventually wear thin.
Note that they might be training on your route
How does it do in the rain?
v12.4.3 still thinks route signs are speed limits
This is probably the biggest issue FSD still has. I love it, but slowing down to 40 on a 75+ mph road because there was a sign for "Route 404" is just something that can't happen. That and stopping for the wrong red light, e.g. one on an overpass running parallel to the road the car is actually on, are the two things making me really nervous for the RoboTaxi. Both of these things have happened to me in my Model 3! It's only that last
"Babe, turn on, new AI DRIVR video just dropped."
I said to my robot girlfriend.
😂
Just moved out of the neighborhood you always drive around in Berkeley / Kensington. So nice to have normal streets now lol those ones are ridiculous even for normal drivers
Great video, as usual! One correction: It's not "miles per intervention", it's "miles between interventions".
My Y running 12.4.3 just learned how to turn into the driveway and line up with the garage but as of yet, won't go in by itself. Maybe I can train it to pull into the garage.
I wouldn't want ChatGPT driving for me. Too many examples of it producing lists of words that end with an n, but actually don't. And then fixing the list incorrectly when that's pointed out.
Love your videos keep it up
Every day we get closer to 2016. I hope it arrives soon.
The last step is to use an LLM to give the car voice feedback so it can listen to the occupant and adjust itself accordingly like a silent cab driver.
This video seems really interesting. Currently I’m looking at buying one of these. It doesn’t seem really like it’s worth $12,000 extra. Maybe the future. It is but right now it seems like you are clearly buying for the future.
FSD works well in city driving. Probably because all the testers drive these areas. They need to spend some time on rural roads where it consistently fails to handle basic traffic scenarios. An unprotected left turn on a 25-35 MPH road is not the same as a 55 MPH road.
More than anything else, they need to work on consistency. I should be able to know how the car is going to react so I am prepared to react. 12.4, for me, is a big step backward from 12.3 in most ways.
In my area v12.4.3 has been quite an improvement over v12.3.6.
For seven years now the improvements in FSD have been huge, awesome, fantastic. Still we are not done. There is not one driverless Tesla to date. 6 years ago I bought a model S with FSD (in Europe), I sold it before I could ever see FSD in action. I no longer drive Tesla (I now drive Xpeng, its FSD is free because it is just as powerful as Teslas and they do not consider that complete). I do like to watch your videos though and keep up to date on the ever lasting experiment. Being frustrated about the past 6 years though the powerful words you use on the progress … hurt. Suggestions for better words: tiny improvements, some improvements but no where near the end product, baby steps as usual. In other words, a little bit less positive would be fair, I think. Unless … you get payed by Elon … in that case, continue with the big words.
It isn't necessarily being payed, but the fact that theTesla influencers might loose access to God, oops I mean Elon, which would destroy their ability to produce "breaking" content.
Very cool. Thanks for
5:20 BEETLE SPOTTED
autonomous vehicles are clearly not happening soon
As good as it is, would you want to insure a car with FSD and no human to supposedly take over in case it screws up? It has to be better than a sober, attentive human in every way, or it isn't ready.
@@AudiTTQuattro2003 Tesla Insurance
Not sure if it was a mic change or something else, but I just noticed your commentary style changed a lot the last couple of videos. It used to be a lot more relaxed, easy to listen to. Now it's way more pronounced but at the same time sounds more "yelling"-like and is tougher on the ears, at least for me.
Try to make the LVCC Loop diver-less first.
A tunnel with no pedestrians, animals, weather or on coming traffic.
I'm still on FSD 12.3.6 and I have been trying to use it for four days. In the neighborhood but yesterday I got out of my comfort zone tried it on a main road kinda like highway like. So when will I get 12.4 and 12.5?
FSD has nothing to do with GPT, why confusing these?
Bro so much work is going on the update for you ah?
Good overview video, thanks!
Thanks for your content
I would like to see 12.4 take exit ramps more naturally…. 12.4 has excessive wheel movements that would be dangerous in slippery or icy conditions.
Hey AI DRIVR, I recently watched your video of, "Can Autopilot drive blinded" Im just curious on what would happen today if you repeated the test.
Thanks for a relatively short video.
Really like summary like this!
Wonderful presentation! You seem to be both super smart and super nice.
How is this legal, it crossed a double yellow 😂
The craziest part is 12.5 will probably release this year, and that's going to be another big shift as it moves FSD's highway stack from hand-coded to AI.
Anyone who has FSD here it would be awesome if you could report your data to FSD community tracker! It is the only way we can continue to follow the improvement trend as it gets better and better!
I would love to see you drive in europe 🙏🏻
You have to have your hands on the steering wheel at all times when you are driving.If you don’t have the car will not drip which to me is necessary.There have been a lot of deaths all over the world because of this and dangerous incident.Not long ago two people where I live died in a crash in a self driving Tesla which drove into a three. Another incident from a Tesla where it suddenly would change lane without the consent of the driver and he had turn the wheel hard to avoid a crash .The first step was the autopilot which now can keep the distance to other cars while driving . Automations can be dangerous as quoted.
Gotta stop you there, just two minutes in.
To say, great intro, beautifully done, tho' I would expect no less from you. Can't wait for the rest, ok onwards...!
Imagine someone uploads those crash cam videos to the AI and it gets pushed in an update...
7:54 I guess they don't have many videos from Teslas driving through fuel stations!
I really wonder what feature engineering they do on location data during training and inference.