OFFICIAL BUTTON WATCH

[AlanSubie4Life]

Autopilot is claimed to be safer when on than off without a "used as designed" caveat.

Of course it's when it's being used as designed - following cautions and warnings.

Because they don't qualify it with "when used as directed" you ASSume that means no driver intervention whatsoever?

When Tesla compiles their statistics, how do they determine which drivers are using Autopilot as designed?

 

[rxlawdude]

"In the 1st quarter, we registered one accident for every 4.19 million miles driven in which drivers had Autopilot engaged. For those driving without Autopilot but with our active safety features, we registered one accident for every 2.05 million miles driven. For those driving without Autopilot and without our active safety features, we registered one accident for every 978 thousand miles driven. By comparison, NHTSA’s most recent data shows that in the United States there is an automobile crash every 484,000 miles."

Where does it say those drivers were operating the vehicles according to the cautions and warnings?

 

[Daniel in SD]

So I guess they can use the FSD cheats and send that vehicle off onto the streets? Seriously, you're arguing for the ridiculous. Because they don't qualify it with "when used as directed" you ASSume that means no driver intervention whatsoever?

Seriously?

Yes, people who drive poorly are included in the statistics when Autopilot is on and when it is off. Seriously.

 

[rxlawdude]

When Tesla compiles their statistics, how do they determine which drivers are using Autopilot as designed?

They don't. That includes morons.

 

[rxlawdude]

Yes, people who drive poorly are included in the statistics when Autopilot is on and when it is off. Seriously.

But that's not your point, is it? Your assertion was that Autopilot associated accidents were not caused by improper use, rather a failure of Autopilot safety.

That's patently wrong. So yes, the statistics include real world data that includes those accidents caused by morons. (I use the term describing those who can't follow directions and remain attentive as drivers.)

 

[AlanSubie4Life]

They don't. That includes morons.

Right, so not sure what you are taking issue with. Obviously this applies to all the categories and would apply to FSD as well when doing a safety comparison.

This started with a comment that they should release the button when FSD is safer than a human, when used as designed. Which does not seem like a great metric. I thought that was the point of the comment but maybe I have lost the thread here.

 

[Daniel in SD]

But that's not your point, is it? Your assertion was that Autopilot associated accidents were not caused by improper use, rather a failure of Autopilot safety.

That's patently wrong. So yes, the statistics include real world data that includes those accidents caused by morons. (I use the term describing those who can't follow directions and remain attentive as drivers.)

Where did I claim that?
Most collisions are caused by driver error whether the driver is using Autopilot or not!

 

[rxlawdude]

Where did I claim that?
Most collisions are caused by driver error whether the driver is using Autopilot or not!

So autopilot contributes what to that. Tesla's statistics clearly show accident incidence is lower with AP on, compared to when it's off.

Or do you disagree with Tesla's stats? In any case, these stats show Teslas with AP engaged (and operated by non-morons) are demonstrably safer than non-Teslas or Teslas where drivers are not using AP.

So getting to the original question, which was the wording of a poll asking about unsupervised accidents on AP/FSD are patently ridiculous.

 

[Daniel in SD]

So autopilot contributes what to that. Tesla's statistics clearly show accident incidence is lower with AP on, compared to when it's off.

Yes, they do show that.

Or do you disagree with Tesla's stats?

No I do not disagree with the stats, I believe Tesla is telling the truth.

In any case, these stats show Teslas with AP engaged (and operated by non-morons) are demonstrably safer than non-Teslas or Teslas where drivers are not using AP.

Yes, possibly similar to the way that it's demonstrably safer to drive with your headlights off because the crash rate is higher with them on. Tesla does not control for when and where people use Autopilot in their statistics so I don't draw conclusions from them. I don't know.
They also use a different definition of accident than the non-Tesla data.
This isn't really relevant to the discussion though and has been argued about many times here.
Tesla does not report any "non-moron" statistics...

So getting to the original question, which was the wording of a poll asking about unsupervised accidents on AP/FSD are patently ridiculous.

I explained repeatedly why the question is not ridiculous (and is a question that Tesla themselves research!). It's just a completely different question than the safety of FSD Beta or Autopilot.

 

[rxlawdude]

I explained repeatedly why the question is not ridiculous (and is a question that Tesla themselves research!). It's just a completely different question than the safety of FSD Beta or Autopilot.

So the statistical rate of unsupervised accidents on FSD is not safety related?

If you're asserting this is related to future iterations of FSD (e.g., L3 or better), I might agree. But that's not what is being measured. What's being measured is with current software and hardware in the real world.

 

[Daniel in SD]

So the statistical rate of unsupervised accidents on FSD is not safety related? Gotcha.

It is not related to the safety of FSD Beta or Autopilot. That's what I said.

It is related to the safety of non-beta (i.e. robotaxi) FSD.

 

[rxlawdude]

It is not related to the safety of FSD Beta or Autopilot. That's what I said.

It is related to the safety of non-beta (i.e. robotaxi) FSD.

I quoted the last public disclosure regarding extant hardware and software, which is exactly what your last sentence states.

In any case, no point in going around and around. We clearly disagree. Let's just leave it there.

 

[Daniel in SD]

I quoted the last public disclosure regarding extant hardware and software, which is exactly what your last sentence states.

In any case, no point in going around and around. We clearly disagree. Let's just leave it there.

I have no idea what you actually disagree with other than the premise of a question asked in a completely different thread.

 

[qdeathstar]

It is not related to the safety of FSD Beta or Autopilot. That's what I said.

It is related to the safety of non-beta (i.e. robotaxi) FSD.

How can there be statistics for something that doesn’t exist?

 

[GeoffEV]

How can there be statistics for something that doesn’t exist?

Exactly! This "Safety" discussion is irrelevant (and off-topic) in that FSD beta is already out on public roads in myriad non-geofenced, unprotected - dare I say: UNSAFE environments?! The only difference is that we, who don't have it, are being denied access to it for no good reason. What makes the 70-some odd lucky sperm club better than us?!

Stop changing the subject, ladies&gents! The topic is the 'Official Button Watch' and, in my recovering/retired engineer's mind, the unequal distribution of it. We should not be requesting "The Button" from Elon, we should be demanding it!

 

[Daniel in SD]

Exactly! This "Safety" discussion is irrelevant (and off-topic) in that FSD beta is already out on public roads in myriad non-geofenced, unprotected - dare I say: UNSAFE environments?! The only difference is that we, who don't gave it are being denied access to it for no good reason.

Stop changing the subject, ladies&gents! The topic is the Official Button Watch and the unequal distribution of it. We should not be requesting "The Button" from Elon, we should be demanding it!

IMO Safety (or perceived safety) is the only plausible explanation for the absence of The Button. I haven't seen anyone propose an alternate explanation but I may have missed it.
I agree that the safety of non-beta FSD is completely irrelevant. When that's available there won't be a button.

 

[GeoffEV]

IMO Safety (or perceived safety) is the only plausible explanation for the absence of The Button. I haven't seen anyone propose an alternate explanation but I may have missed it.
I agree that the safety of non-beta FSD is completely irrelevant. When that's available there won't be a button.

"Perceived safety"! Yes!! So, we the unwashed, are the victims of Tesla lawyers and PR? So, let's push for the rollout of The Button using our state certified driver's records. Let's put the power of our goobermints' knowledge of our driving records to work for those of us with zero points on our drivers' licenses.

 

[helvio]

IMO Safety (or perceived safety) is the only plausible explanation for the absence of The Button. I haven't seen anyone propose an alternate explanation but I may have missed it.
I agree that the safety of non-beta FSD is completely irrelevant. When that's available there won't be a button.

I have. They don't want to deal with the amount of feedback that it would generate.

 

[Daniel in SD]

I have. They don't want to deal with the amount of feedback that it would generate.

They've already developed industry leading technology to avoid customer interaction and feedback though.

 

[helvio]

They've already developed industry leading technology to avoid customer interaction and feedback though.

I'm not saying that's a fact. It's a possibility. No one here knows the facts. Feedback happens not only through Email and the camera feedback button, but also on social media, especially YouTube.