Live data from Hacker News

Apple engineer killed in Tesla crash had previously complained about autopilot

kqed.org

281–290 of 919 posts

Re: Apple engineer killed in Tesla crash had previously complained about autopilot

#281

Earlier quoted context omitted.

I don't know where you got these papers but you clearly got an awful sampling and I don't think you're giving computer vision a fair evaluation. Granted, we're not quite ready for self driving, but there's no question that the neural network subfield of ML has absolutely exploded in the last 5-10 years and is bursting with productionizable, application ready architectures that are already solving real world problems.

There is 40-50% drop in precision with state of the art results if test images are ill-formed. The imagenet dataset used is far from ready for real world use cases. A bunch of IBM and MIT researchers are trying to fix this - https://objectnet.dev/

As in, "only works in great visibility on a perfectly spherical road"? That does seem an appropriate summary.

Re: Apple engineer killed in Tesla crash had previously complained about autopilot

#282
post #19

Earlier quoted context omitted.

GM calls theirs Super Cruise which is true hands-free. https://www.youtube.com/watch?v=RxeK0F-D3gg

yeah, as long as you're staring straight ahead. It has a camera that watches your face. Otherwise they're the same. Except Tesla's keeps getting better with each update and GM stays the same until you take it in to the shop...

> GM stays the same until you take it in to the shop...

Over the air software upgrades are supported by GM.

"When the work is done, an over the air software update will turn it back on."

https://www.wired.com/story/cadillac-super-cruise-ct6-lidar-...

Re: Apple engineer killed in Tesla crash had previously complained about autopilot

#283
post #269

Earlier quoted context omitted.

Regardless of what autopilot means to pilots, it means something else to drivers.

Why? Autopilot is a term borrowed directly from aviation.

Most drivers aren't pilots. They don't know what autopilot actually does. They expect it to drive the car for them, and they probably think autopilot does the same on planes. It was a mistake for Tesla to borrow the term.

Re: Apple engineer killed in Tesla crash had previously complained about autopilot

#284
post #19

Earlier quoted context omitted.

GM calls theirs Super Cruise which is true hands-free. https://www.youtube.com/watch?v=RxeK0F-D3gg

I like the idea of super cruise -- it seems like a lidar scan of the freeway would have prevented Huang's death. But, obviously the freeway changes over time so I wonder how frequently GM re-scans the freeways. Lidar is expensive of course, but it also seems so obvious that the lidar needs to be onboard the vehicle.

Not sure but apparently they detect road works and are able to update the maps remotely.

Re: Apple engineer killed in Tesla crash had previously complained about autopilot

#285

Earlier quoted context omitted.

Can you cite any sources that AI fails harder? I'm not exactly sure what you are referring to.

I believe he's alluding to the nature of the accidents. They're high intensity events which are more likely to be fatal (speeds were ~70mph). They're not fender benders when the autopilot fails.

Auto-pilot that drives above the speed limit ought to indicate to you its scope.

Still seems like people treat auto-pilot like auto-drive and die as a result.

Thats not a tech fail imho

Re: Apple engineer killed in Tesla crash had previously complained about autopilot

#286
post #266

The headline alone should show, what is odd with this case. * Tesla clearly states, that the autopilot is a driving aid and should not be used unobserved. * The engineer complained about the autopilot not working perfectly at the very place of crash. (Quite understandeable when seeing the state of the road and mostly absence of line markings in videos of the crash site). * Evidence seems to show that prior to the cra…

> Tesla clearly states, that the autopilot is a driving aid and should not be used unobserved.

From Tesla's own promotional video[1]:

> The person in the driver’s seat is only there for legal reasons. He is not doing anything. The car is driving itself.

[1] https://www.tesla.com/videos/autopilot-self-driving-hardware...

Re: Apple engineer killed in Tesla crash had previously complained about autopilot

#288
post #283

Earlier quoted context omitted.

Why? Autopilot is a term borrowed directly from aviation.

Most drivers aren't pilots. They don't know what autopilot actually does. They expect it to drive the car for them, and they probably think autopilot does the same on planes. It was a mistake for Tesla to borrow the term.

You're purely speculating on people's knowledge:

> They don't know

> They expect

> they probably think

The alternative is some people are terrible and there will always be people that abuse nice things on purpose.

They know what they are doing, that's why they are trying to trick the warning systems.

These are the same people on the phones and merging without looking in normal cars.

All of that is speculation, but imo it's more likely than yours.

Re: Apple engineer killed in Tesla crash had previously complained about autopilot

#289

Earlier quoted context omitted.

You know, it’s almost like we’d benefit from having an autopilot system. I mean, think about it: people drive drunk, or tired, or get in fights with their mom on the phone, or have a crying baby demanding attention, or any number of other things like that. Wouldn’t it be awesome if we had something to help with that? Like, it doesn’t even have to do everything, just most of it, and keep you from running into shit. Th…

You can wish all we want but you don’t have an auto pilot system. The Tesla is auto pilot up until you actually do what their marketing (and fans) claim it does and get in an accident. Then both Tesla and the fanboys come out of the woodwork denying that it was ever intended to be used hands off even for a second. In fact Tesla goes way out of their way, presenting telemetry from the car itself “proving” the driver w…

You are focused upon strawmen and arguing in bad faith.

It is a scientific fact that AP is safer than not having it. Statistically, you are less likely to get in an accident if you have it. Anecdotally, I have witnessed this with my own eyes, having had it take drastic action before I was even aware of the danger.

I’m done with you specifically, though. I’ve come to believe that anyone who uses the word “fanboy” is almost never someone you can have a productive conversation with.

Re: Apple engineer killed in Tesla crash had previously complained about autopilot

#290
The thing about self-driving cars is that even though we may get to a point where they are statistically better at driving than humans, the reasons they crash will be unrelated to the driver. This means that humans, thanks to modern life, are even further removed from any concept of self-determination.

Being a responsible person and not driving while drunk, or driving in a reckless manner will no longer improve your chances of not being seriously hurt in a car crash. While driving on the roads already removes a lot of self-determination due to the fact that you can't control all the high-speed objects flying around you, there was at least some degree of improvement possible by being responsible.

Once self-driving cars are available, whether you live or die will be up to random chance. Some people will be better off due to the self-driving cars, and others will be worse off, depending on if they were above the average or below in terms of their driving behaviour.

Post reply on HN