They operate in Los Angeles on the city streets in a square between Marina Del Ray and West Hollywood.
They can definitely do better when taking left turns. I've seen situations where Waymo depends on the oncoming drivers to slow down.
New Study: Waymo is reducing serious crashes and making streets safer
41–50 of 413 posts
Re: New Study: Waymo is reducing serious crashes and making streets safer
#42Waymo says Waymo is safe. https://www.latimes.com/california/story/2024-05-16/woman-ge...
Tens of thousands of Americas are killed every year in car crashes. The existence of a case or some cases where a self-driving car caused injury has zero value. What matters is the rate of cases per mile driven.
The reason they investigate every airplane crash for a root cause is not necessarily because the number of deaths tilts the statistics, it's because they are highly complicated mechanical and electrical machines where there is always the chance for a systematic fault that will doom many further planes if not found and rectified.
The same of course applies to self-driving cars; they are literally cars driven by software, of course you need to do a root cause investigation every time to rule out that it's not a bug in the software that will kill another person (and many after) when the next car happens to go down that rare branch of the system. It's embarrassing to see that the people that call themselves engineers at these companies have not held their work to this standard, and are instead publishing glossy brochures making whacky statistical arguments.
Re: New Study: Waymo is reducing serious crashes and making streets safer
#43How do AI cars fare with those instinctual "decide which way to swerve in a split-second" scenarios that come up maybe once every 10-20 years over the course of a driver's history?? It's happened to me about 2-3 times and I've always made the "(assumedly) correct" split second swerve decision. Wondering if that is a "human/instinctual" skill and if AI cars do just as well or better, or perhaps not as good? I don't ha…
Probably better? 'Who' can process more data faster is likely impossible to answer, but (e.g.) Waymo can train on those scenarios and have way more 'experience' than any individual driver who's seen it once every 10-20 years of their driving career.
Re: New Study: Waymo is reducing serious crashes and making streets safer
#44Earlier quoted context omitted.
The US drives about 3.2 trillion miles per year. Waymo has 56.7 million miles over several years. Their percentage data is essentially useless. 50% of fatalities involve alcohol or drugs and are often single vehicle accidents. 25% involve youth or inexperience. 15% involve motorcycles. 15% involve pedestrians. What I really need to see is a complete breakdown of every accident a Waymo has had. Then I can start to com…
> The US drives about 3.2 trillion miles per year. Waymo has 56.7 million miles over several years. Their percentage data is essentially useless. No, that's not how statistics works. The percentage data's accuracy depends mainly on the number of incidents recorded (and somewhat on the rate of incidents). But the percentage of the whole is completely irrelevant. If you are basing something on 10 incidents but it's 50%…
> If you are basing something on 10 incidents but it's 50% of the total, it's still terrible accuracy.
The ratio of 3.2 trillion to 56.7 million, which is already incredibly generous to Waymo's position, is 5 orders of magnitude in difference. So any calculations from Waymos data are going to be insanely inaccurate and not something you can extrapolate from.
The main, and most obvious case, evidenced by this, is Waymo does not operate where snow falls. Human beings do.
We're missing so much of the picture I don't think you can say Waymo's are 75% less accident prone, or 80% less likely to hit a pedestrian. Those are just nonsense numbers.
Re: New Study: Waymo is reducing serious crashes and making streets safer
#45How do AI cars fare with those instinctual "decide which way to swerve in a split-second" scenarios that come up maybe once every 10-20 years over the course of a driver's history?? It's happened to me about 2-3 times and I've always made the "(assumedly) correct" split second swerve decision. Wondering if that is a "human/instinctual" skill and if AI cars do just as well or better, or perhaps not as good? I don't ha…
> those instinctual "decide which way to swerve in a split-second" scenarios that come up maybe once every 10-20 years
The correct answer is almost always to hit the brakes. Not to swerve. And Waymo will hit the brakes earlier than you or me.
Re: New Study: Waymo is reducing serious crashes and making streets safer
#46That's great news. I wonder how much insurance rates will go down when autonomous vehicles get popular. Seems like the liability portion could go way down.
It's a 3x effect for insurance agencies: 1) Google/Waymo doesn't buy insurance for their vehicles 2) Waymo vehicles are much less likely to cause accidents 3) Waymo vehicles actively avoid being hit and calm traffic overall
I'm not super familiar with this space, or laws where Waymo is active, what do you mean Google/Waymo don't buy insurance for their vehicles?
Is having insurance not legally required? Do they just pay out when there's an accident where they injure someone?
Re: New Study: Waymo is reducing serious crashes and making streets safer
#47Earlier quoted context omitted.
The problem is that Waymos can't replace cars (or even Ubers!) since they depend on human oversight and problem-solving. Even at a 20:1 ratio it is not feasible for autonomous vehicles to employ 10m people to account for every American on the road.
Americans only spend 3% of their time driving, so it's maybe closer to 300k than 10m. Still not scalable at that though so the key will be driving that 20:1 up, which seems very possible given enough time.
What determines whether 300k is scalable ?
Re: New Study: Waymo is reducing serious crashes and making streets safer
#48Earlier quoted context omitted.
Tens of thousands of Americas are killed every year in car crashes. The existence of a case or some cases where a self-driving car caused injury has zero value. What matters is the rate of cases per mile driven.
The US drives about 3.2 trillion miles per year. Waymo has 56.7 million miles over several years. Their percentage data is essentially useless. 50% of fatalities involve alcohol or drugs and are often single vehicle accidents. 25% involve youth or inexperience. 15% involve motorcycles. 15% involve pedestrians. What I really need to see is a complete breakdown of every accident a Waymo has had. Then I can start to com…
You'll be happy to know, then, that the NHTSA publishes the data you're looking for at https://www.nhtsa.gov/laws-regulations/standing-general-orde..., in addition to Waymo's own reporting in their safety hub. For me personally, I find summary data to be more informative than a pile of individual reports, but I hope your comparison goes well!
Re: New Study: Waymo is reducing serious crashes and making streets safer
#49For cycling and walking in SF, the Waymos are optimal. If there's one at the front of the line at the red light I know I won't be right hooked as I go straight in the bike lane. Very well-behaved. I am so glad they're here. Some concentration of them actually ruin crazy drivers' ability to do damage because they set the speed limit and won't go through reds.
Re: New Study: Waymo is reducing serious crashes and making streets safer
#50Earlier quoted context omitted.
Tens of thousands of Americas are killed every year in car crashes. The existence of a case or some cases where a self-driving car caused injury has zero value. What matters is the rate of cases per mile driven.
For which any such system must be assumed unsafe until proven safe and the burden of proof lies with the manufacturer to present sound, robust, transparent, third-party audited evidence. You do not get to counter-argue: “What matters is the rate of cases per mile driven” without actually presenting that number with supporting evidence. Otherwise the only sound conclusion is the default presumption of non-safety. In t…
I'm not sure what you're getting at here. By all accounts, Waymo operates more safely than a human driver. It's operated tens of millions of rides in several major cities over several years, hence all the useful data. There is absolutely nothing stopping the continued expansion of self-driving vehicles, Waymo or otherwise, and there are myriad factors moving it forward rapidly.