New Study: Waymo is reducing serious crashes and making streets safer

As someone who is often on SF city streets without a car - I bike and run a lot - I absolutely love Waymo. I am continuously seeing human drivers cut me off, perform illegal maneuvers (i.e. run red lights when I&#x27;m going through a crosswalk), and break various other traffic laws. All these things genuinely put people in danger. Just the other day, a guy started running a &quot;no right turn on red&quot; lane in SF, and when I pointed it out to him he floored his car - through the red - right in front of me and laughed at me as he sped away. To say nothing of all the times when cars will honk or give me the finger for doing normal things on a street, like walking on a crosswalk.<p>Waymo is like the most courteous, respectful driver you can possibly imagine. They have infinite patience and will always take the option which is the safest for everyone. One thing which really impressed me is how patient they are at crosswalks. When I&#x27;m jogging, a Waymo will happily wait for me to cross - even when I&#x27;m 10 feet away from even entering the crosswalk! I don&#x27;t know if <i>I</i> even have that much patience while driving! I&#x27;ve had a number of near misses with human drivers who don&#x27;t bother checking or accelerate for no reason after I&#x27;m already in the crosswalk. Can you imagine a Waymo ever doing that?<p>If I see a Waymo on the street near me I immediately feel safer because I know it is not about to commit some unhinged behavior. I cannot say enough good things about them.

This will surely get some skepticism as it&#x27;s a Waymo study, but it&#x27;s nice to see a real‐world dataset this large at 56M miles. An 85 % drop in serious‐injury crashes and 96 % fewer intersection collisions is a strong signal that Level 4 ADS can meaningfully improve safety in ride-hail settings. Still curious about how much of that comes from operational design versus the core autonomy, but it’s a big leap beyond “novelty demo.”<p>Really excited for autonomy to become more and more common place. People drive more and more like distracted lunatics these days it seems

Literally riding in a Waymo right now in Los Angeles.<p>IMO they already won. The amount of stupid things you see people do here while driving is astonishing, so many people are not paying attention and looking at their phones.<p>I used an Uber on the way here and the car was dirtier while the service was identical (silent ride, got me where I needed to go.)<p>I’ve also been stuck in a Waymo that couldn’t figure out its way around parked buses, so they have edge cases to improve. But man does it feel like I’m living in the future…

Was genuinely impressed when I took my first Waymo, not only for comfort, but the small microdecisions it made as a driver. As a person whose lost a parent to a sleepy driver, and a victim to 2 texting drivers, I welcome AI driving revolution.

Anecdotally, as someone who bikes a lot in SF, Waymo&#x27;s are a lot safer than human drivers simply because they follow the letter of traffic laws. Stopping at stop signs, waiting for pedestrians to clear the box, following the posted speed limits, etc.

This is great and there’s another area of influence that I’ve heard other traffic engineers discuss: platoon pacing. A platoon is the word that traffic engineers give to a group of cars traveling together. A platoon is most explicitly visible on a corridor with signals timed for a green wave, but occurs in many other contexts.<p>Human drivers often race when in a platoon— not even on purpose it’s just an instinct to go as fast or faster than other cars which has a feedback effect to increase platoon speed.<p>Waymos, following the exact speed limit, don’t do this. On 1 lane streets they literally set the platoon pace to the legal speed limit.<p>The effect of this is hard to study and quantify but it’s a real and positive impact of self driving cars on city streets. Haven’t seen research on this topic yet.

I saw a Waymo stop at a crosswalk last night in the dark where there was a person standing there waiting to cross that I don&#x27;t think I would have seen. The person was not standing out in the road, they were standing there patiently waiting to make sure the car actually stopped since it was dark. I was really impressed! I don&#x27;t think I&#x27;m a reckless or impatient driver, but I think the Waymo&#x27;s are probably better at driving than I am. I know I prefer the Waymos to the human drivers I typically see on the road.

For those that are interested, our Safety Research team also makes the underlying data available for download:<p><a href="https:&#x2F;&#x2F;waymo.com&#x2F;safety&#x2F;impact&#x2F;#downloads" rel="nofollow">https:&#x2F;&#x2F;waymo.com&#x2F;safety&#x2F;impact&#x2F;#downloads</a><p>(I don&#x27;t work on that team, but I&#x27;ve noticed a few comments that would be better served with their own analysis on top of the available data)

For what it&#x27;s worth, I have exclusively used Waymo in SF ever since my last Uber driver was smoking weed while driving. I just don&#x27;t want to deal with the human variable anymore.

For cycling and walking in SF, the Waymos are optimal. If there&#x27;s one at the front of the line at the red light I know I won&#x27;t be right hooked as I go straight in the bike lane. Very well-behaved. I am so glad they&#x27;re here. Some concentration of them actually ruin crazy drivers&#x27; ability to do damage because they set the speed limit and won&#x27;t go through reds.

How do AI cars fare with those instinctual &quot;decide which way to swerve in a split-second&quot; scenarios that come up maybe once every 10-20 years over the course of a driver&#x27;s history?? It&#x27;s happened to me about 2-3 times and I&#x27;ve always made the &quot;(assumedly) correct&quot; split second swerve decision. Wondering if that is a &quot;human&#x2F;instinctual&quot; skill and if AI cars do just as well or better, or perhaps not as good? I don&#x27;t have any evidence backing this but my gut tells me these scenarios are something that a human driver would handle better than AI.

In Holland we have this saying (based on an old commercial): &quot;We from WC Duck recommend... WC Duck&quot;. (Which is a toilet cleaning product)<p>It was a funny ad at the time. Unfortunately based in reality more and more these days.

So, what’s it gonna take for Waymo to start selling retrofit kits for existing cars?<p>If a $10,000 investment reduces the chances of a serious accident by 90%, the corresponding reduction in insurance rates might have a payoff within a few years. Especially if adoption starts to push rates up for customers who don’t automate. I can’t take a taxi everywhere, but I’d sure like it if my car drove me everywhere and did a better job than me at it too.

Waymo is part of my investment thesis as to why Google is undervalued

We don&#x27;t have waymo in France and I am not sure it wrote work well.<p>Streets in France are full of entitled people. Drivers, bikers (both bikes with pedals and the ones with an engine) and pedestrians.<p>Everyone thinks that they have all the rights but ultimately some kind of order emerges from the chaos.<p>Pedestrians will walk on red lights, but are also careful.<p>Cars will park anywhere, but usually in a way that is just really frustrating but not blocking.<p>Bikes will slalom, but to a point.<p>This does not always work, but it is so much driven by culture that somehow we are statistically alive when moving outside.<p>Personally I hate it with all my heart. I dream of the dystopian world where everyone will follow the rules.

Well yeah, they stack the deck in their favor. They avoid very hilly roads. They don’t yet go on the freeway. And they don’t get drunk or tired or distracted.<p>That being said I’d like to see how a typical “good” driver compares vs the average. Someone who doesn’t speed or get duis and gets plenty rest.

Can we mention that drivers using Tesla’s FSD are 5x less likely to be in an accident than human drivers?<p>Or all things Elon bad?

i am suspicious of all the anti-human-driver comments and all the dismissal of any concerns about Waymo in the comments here.<p>I am not convinced that public testing of such services is safe, let alone commercial service. One cannot punish a self driving vehicle in any meaningful sense. Corporate incentives vs the public commons, is a general concern that cannot be sweettalked away.<p>The metaphors about human drivers recording you also seem like reductio ad absurdum.<p>puff pieces like this should not be well received on HN or it discredits any pretence at separation of concerns with regards to HN and ycomb.

That&#x27;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.

&gt;The research finds that, compared to human benchmarks over 56.7 million miles and regardless of who was at fault, the Waymo Driver had [list of better than human stats]<p>Well considering this sensor package...<p>&gt;With 13 cameras, 4 lidar, 6 radar, and an array of external audio receivers (EARs), our new sensor suite is optimized for greater performance...it provides the Waymo Driver with overlapping fields of view, all around the vehicle, up to 500 meters away, day and night, and in a range of weather conditions.[0]<p>...I would hope it is considerably better than humans who are limited to a sensor suite of two cameras and two lower-case ears.<p>[0] - <a href="https:&#x2F;&#x2F;waymo.com&#x2F;blog&#x2F;2024&#x2F;08&#x2F;meet-the-6th-generation-waymo-driver&#x2F;" rel="nofollow">https:&#x2F;&#x2F;waymo.com&#x2F;blog&#x2F;2024&#x2F;08&#x2F;meet-the-6th-generation-waymo...</a>

Waymo per their track record I trust and am excited to try &#x2F; use it once it comes to DC or another city close to DC.<p>All the other big names that are no longer around... their tech was dangerous and definitely not ready for prime time. Their tech and focus seemed all about making all involved wealthy or wealthier.

Once self driving cars take over and fill the streets with gridlock autonomous traffic, the only way to get around in a city is going to be by bike.<p>Its great to here their algorithms are good for cyclists, a better solution is to keep investing in infrastructure that separates cyclists completely from cars.

They operate in Los Angeles on the city streets in a square between Marina Del Ray and West Hollywood.<p>They can definitely do better when taking left turns. I&#x27;ve seen situations where Waymo depends on the oncoming drivers to slow down.

This article has a huge conflict of interest. I would like more independent data.

While driving a car, it is possible to do something, even on accident, that can land a person in jail. These crimes do not have the option of paying a fine in lieu of prison time.<p>A &quot;self-driving&quot; car can cause the same accident but gain advantages over a human driver that the person ultimately responsible is no longer held to the same set of laws.<p>This seems to undermine foundations of law, placing the owners of those assets into a different legal category from the rest of us.

reminder that if waymo type cars replaced human driven cars and cut the deaths to zero. (perhaps a big if)<p>Then it&#x27;d be like finding a cure for cancer, for people aged 0 - 40, who die as much in auto accidents as they do of cancer

I&#x27;ve noticed that nearly ANY criticism here is grayed out.<p>HN has a credibility problem here.

Why is Waymo waiting so long to roll out more broadly?

source waymo.com???

A few weeks ago we were in Santa Monica on vacation. This was my first time seeing Waymo vehicles in the wild. We did see one blatantly run a red light, well after the light had changed. Fortunately no car was crossing at the same time. It wasn’t like the sun was somehow obscuring the traffic light at that time. I was surprised it did something so unsafe.

what&#x27;s up with all the waymo sycophantry here? choruses repeating the same claims &quot;its ok for waymo to enter the bike lane&quot; &quot;yeah, waymo is right&quot; &quot;yes, good waymo entering the bike lane because reasons&quot;<p>this is blatantly obvious and unacceptable.<p>the site rules prohibit accusations of astroturfing but that is precisely what is going on here.<p>precisely no sf programmers were convinced, either.<p>this site had better be concerned with future legitimacy and not being seen as a puppet of specific corps like waymo.

Waymo is using public roads 25&#x2F;7 for automating profit and not paying fair taxes.

While Waymo is good for safety, it is constantly collecting data not just from users, but from everyone nearby. The vehicles essentially function as mobile surveillance devices, recording bystanders without consent, with no clear policies on data use, retention, or oversight.

One way to inteprent their data is that Waymo is LESS efficient protecting cyclists and motorcyclists compared to pedestriants. As a motorcycle rider I hope they will work to fix that gap.

It&#x27;s a great start, but also shows how driving is not a navigation challenge as much as it is a socialization challenge. If I drove right up to some responders dealing with an emergency and repeatedly refused to stop then I could be in big legal trouble and could end up with fines and possibly an arrest. This has not happened with Waymo vehicles because there is legal ambiguity about who is responsible in such cases. Realistic analysis shows these vehicles have a driving record similar to that of overconfident teens which is worrisome.

Waymo says Waymo is safe. <a href="https:&#x2F;&#x2F;www.latimes.com&#x2F;california&#x2F;story&#x2F;2024-05-16&#x2F;woman-gets-millions-after-getting-dragged-by-self-driving-taxi-in-san-francisco" rel="nofollow">https:&#x2F;&#x2F;www.latimes.com&#x2F;california&#x2F;story&#x2F;2024-05-16&#x2F;woman-ge...</a>