Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Tesla Vehicle Safety Report

tesla.com

Post reply on HN
Print view View original

81–90 of 94 posts

Re: Tesla Vehicle Safety Report

#81

Post by chroma »

FireBeyond wrote: post #71
chroma wrote: post #62

Earlier quoted context omitted.

That's what Tesla is counting. From the methodology section of the safety report: > To ensure our statistics are conservative, we count any crash in which Autopilot was deactivated within 5 seconds before impact, and we count all crashes in which the incident alert indicated an airbag or other active restraint deployed.

Yeah, the "no active restraint deployed" thing is horseshit, and Tesla knows it, too. I go to manufacturer training sessions on these systems maybe once a year as a firefighter-paramedic, and there are multiple criteria for deployments these days - you can have a significant collision and if the system determines its better for the occupants not to deploy, then it doesn't. According to Tesla, that's not an accident.…

I was responding to the point about counting crashes after autopilot disengagement, not the criteria for what counts as a crash.

Re: Tesla Vehicle Safety Report

#82

Post by culi »

ajross wrote: post #78
culi wrote: post #50

Earlier quoted context omitted.

the National Highway Traffic Safety Administration, New York Times, and even published research[0] have all been calling Elon Musk out for years now, yet his fans will still believe him despite them withholding data [0] https://engrxiv.org/preprint/view/1973/3986

This is the gallop part. I called you out on your first set of links (one of which legitimately hoodwinked a commenter here!) and instead of defending those arguments you're pivoting to another. I won't engage here either, except to point out that this isn't "published research" (it literally tells you that on the first page). The clear and inarguable truth is that the Tesla fleet is pushing three million cars now. T…

This study is from one of the original links I shared...

You accused me of gish gallopping. I've since slowed down and focused on specific events. See my other comments about the battery swapping scam for example.

Please provide some actual data proving these cars are safer than average. Tesla certainly hasn't been able to. And research, like the one I just linked which you refused to acknowledge, definitely does not back up Elon's claims

Re: Tesla Vehicle Safety Report

#83

Post by ajross »

culi wrote: post #82
ajross wrote: post #78

Earlier quoted context omitted.

This is the gallop part. I called you out on your first set of links (one of which legitimately hoodwinked a commenter here!) and instead of defending those arguments you're pivoting to another. I won't engage here either, except to point out that this isn't "published research" (it literally tells you that on the first page). The clear and inarguable truth is that the Tesla fleet is pushing three million cars now. T…

This study is from one of the original links I shared... You accused me of gish gallopping. I've since slowed down and focused on specific events. See my other comments about the battery swapping scam for example. Please provide some actual data proving these cars are safer than average. Tesla certainly hasn't been able to. And research, like the one I just linked which you refused to acknowledge, definitely does not…

> Please provide some actual data proving these cars are safer than average. Tesla certainly hasn't been able to.

But... they have. That's exactly the link we're discussing. You don't want that to be true, so you're digging around trying to pretend via a flood of bad evidence that they're somehow "lying" in that data.

But... what if they aren't? Again, three million cars are an awfully big signal to try to spin. Isn't the Occam's Razor interpretation (also jives with my personal experience with the product, FWIW) that... they're safe?

Re: Tesla Vehicle Safety Report

#84

Post by bryanlarsen »

dbt00 wrote: post #74
bryanlarsen wrote: post #30

Earlier quoted context omitted.

Better a false positive (phantom braking) than a false negative (not braking when it should). I imagine they're all tuned with this in mine.

Are you kidding me? One of the several times autopilot tried to kill me was when it randomly tried to flat stop on I-80 with a semi driver following me. No way he could have stopped in time if I hadn't disengaged and slammed on the accelerator. (Absolutely no reason to stop and no weather or afternoon sun for confusion. My best guess is that flaky vision systems thought a semi truck was entering my lane, but I really…

But it didn't kill you. If you had a false negative, that means it failed to stop when it should have, and that almost definitely would have caused an accident. False negatives are more dangerous than false positives. Both can be dangerous, but false negatives are worse.

All AEB's from all manufacturers have a certain amount of phantom braking, but non-Tesla AEB are well studied and definitely save lives.

Re: Tesla Vehicle Safety Report

#85

Post by culi »

ajross wrote: post #83
culi wrote: post #82

Earlier quoted context omitted.

This study is from one of the original links I shared... You accused me of gish gallopping. I've since slowed down and focused on specific events. See my other comments about the battery swapping scam for example. Please provide some actual data proving these cars are safer than average. Tesla certainly hasn't been able to. And research, like the one I just linked which you refused to acknowledge, definitely does not…

> Please provide some actual data proving these cars are safer than average. Tesla certainly hasn't been able to. But... they have. That's exactly the link we're discussing. You don't want that to be true, so you're digging around trying to pretend via a flood of bad evidence that they're somehow "lying" in that data. But... what if they aren't? Again, three million cars are an awfully big signal to try to spin. Isn'…

I can't tell if you forgot a there but just in case you're sincere...

I just paid $5k on a used extended van. One of the most dangerous cars you can buy. Obviously I don't expect this to be any safer than a $100k Tesla.

Similarly, I find any comparison between the "average car" and a Tesla's crash ratings to be completely useless. How do we know that other luxury cars around that price point don't actually do much much better than tesla does? The "data" provided is insincere at best. It's also not actual data. They're just telling us some numbers. They've consistently refused to provide any actual data

This OP also again makes the claim of the "lowest probability of injury of any car by NHTSA" which the NHTSA has repeatedly asked them to stop saying.[0] The NHTSA doesn't rank cars by probability of injury at all

Remember when they said the Tesla S earned 5.4 stars and NHTSA had to make a public statement that it does not award more than 5 stars? lol. Yet another lie

[0] https://krcrtv.com/features/auto-matters/nhtsa-to-tesla-stop...

Re: Tesla Vehicle Safety Report

#86

Post by culi »

CamperBob2 wrote: post #27
bagels wrote: post #17

Earlier quoted context omitted.

Autopilot does emergency braking for nonexistent obstacles. I'm sure it's contributing to rear end collision risk. I stopped using it for this reason.

Every autonomous braking system has that problem, including the one in my 2020 Macan. They are dangerously-flaky hacks. As of 2025 they are mandated by law.

Someone else on HN commented this the other day. RSymons RSEV did a comparison[0] of the self-park feature of Tesla's compared to three other cars. All cars nailed the spots they tested flawlessly while the Tesla just straight up didn't see most of the spots. When they moved the cars further and further away so that the spot became more obvious it eventually did see the spot but parked over a foot from the curb

Tesla seems to have a lot of software issues when it comes to "smart" features like these. I wouldn't be surprised if they are indeed worse than the competition, but I'd love to see a similar comparison

[0] https://www.youtube.com/watch?v=nsb2XBAIWyA

Re: Tesla Vehicle Safety Report

#87

Post by culi »

CamperBob2 wrote: post #29
cyrux004 wrote: post #28

Earlier quoted context omitted.

Apparently not as common as Tesla. You should read about "phantom braking" issues on Tesla , especially after they removed the radar

Yeah, I can't even imagine what it would be like to implement a system like that without radar. Tesla's approach would make a good Most Interesting Man In the World meme. "I don't always tackle the hardest engineering problems in human history, such as self-driving cars. But when I do, I tie one hand behind my back, close one eye, and consume some alcohol or drugs first." One episode per 3000 miles is too many, and t…

Musk's obsession with not relying on radar and using only cameras as sensors is nuts and likely a big reason why competitors systems perform so much better. But it seems Tesla has finally backtracked on this

https://electrek.co/2022/06/08/tesla-files-use-new-radar-con...

Re: Tesla Vehicle Safety Report

#88

Post by culi »

bryanlarsen wrote: post #84
dbt00 wrote: post #74

Earlier quoted context omitted.

Are you kidding me? One of the several times autopilot tried to kill me was when it randomly tried to flat stop on I-80 with a semi driver following me. No way he could have stopped in time if I hadn't disengaged and slammed on the accelerator. (Absolutely no reason to stop and no weather or afternoon sun for confusion. My best guess is that flaky vision systems thought a semi truck was entering my lane, but I really…

But it didn't kill you. If you had a false negative, that means it failed to stop when it should have, and that almost definitely would have caused an accident. False negatives are more dangerous than false positives. Both can be dangerous, but false negatives are worse. All AEB's from all manufacturers have a certain amount of phantom braking, but non-Tesla AEB are well studied and definitely save lives.

GP gave an obvious example of a false positive. I don't why them thinking fast enough to save their life neglects the point...

Re: Tesla Vehicle Safety Report

#89

Post by bagels »

bryanlarsen wrote: post #35
CamperBob2 wrote: post #29

Earlier quoted context omitted.

Yeah, I can't even imagine what it would be like to implement a system like that without radar. Tesla's approach would make a good Most Interesting Man In the World meme. "I don't always tackle the hardest engineering problems in human history, such as self-driving cars. But when I do, I tie one hand behind my back, close one eye, and consume some alcohol or drugs first." One episode per 3000 miles is too many, and t…

Are you referring to the "gentle slow down" behavior that Autopilot often does or the "slam on the brakes as hard as you can" behavior of Automatic Emergency Braking? I definitely see the former more than once per 3000 miles but it doesn't bother me. I've never seen the latter.

On my car, all of these occurred frequently.

It definitely is erratic with the speed, speeding up and slowing down slightly, which is annoying, but not so dangerous.

The dangerous parts are driving home from work, 6 miles, at least two very hard brakes each trip with no cars ahead of me (I only made a few trips like this before disabling it).

Automatic emergency braking (with the loud beeps and red message on the screen) on two lane highways, over crests, a myriad of other circumstances.

Re: Tesla Vehicle Safety Report

#90

Post by alphabettsy »

seeekr wrote: post #34
Imnimo wrote: post #26

I do believe that Teslas are quite safe despite the high-profile failures, but I'm very unconvinced by a stat that compares accidents per mile using autopilot to accidents per mile not using autopilot. Those are not comparable miles - people don't uniformly use autopilot across all driving conditions. Imagine you had a subsystem that did absolutely nothing, but people turned it on when driving on empty highways. It's…

I don't disagree that it's not the best of comparisons (and I wonder if a better one could be imagined + implemented...). But still, it's not like we turn on Autopilot/FSD only "on empty highways", far from it! Certainly it's a tool where the user needs to learn its strengths and weaknesses and use it accordingly, but it is useful in so many more situations than not, that it's also not a terrible or meaningless compa…

> The alternative would often be to slow down by a significant amount

You should probably slow down.

Post reply on HN