Today, Explained - Riding in Cars with Robots

Episode Date: July 27, 2022

The data is in on autonomous cars: They are crashing, but they're still doing a lot better than regular cars driven by humans. The Verge’s Andrew Hawkins and Vox’s Marin Cogan take the wheel. This episode was produced by Jon Ehrens, edited by Matt Collette, fact-checked by Laura Bullard, engineered by Paul Mounsey, and hosted by Sean Rameswaram. Transcript at vox.com/todayexplained   Support Today, Explained by making a financial contribution to Vox! bit.ly/givepodcasts Learn more about your ad choices. Visit podcastchoices.com/adchoices

Computer, drive me to work. Okay, Sean, driving you to DuPont Circle in Washington, D.C. Computer, you know, I've never asked you this before, but are you any good at driving, statistically speaking? Of course, Sean. I make far fewer mistakes than humans do. On balance, you people are terrible at driving. You people. Wow. But, you know, computer, I hear you've been in a bunch of accidents recently. You know, we should do a show about this. Ahead on Today Explained, the data's in on automated vehicles, on driver-assisted vehicles, and they are crashing, but they're still nowhere near

as deadly as regular cars driven by regular people. To who are you speaking? The all-new FanDuel Sportsbook and Casino is bringing you more action than ever. Want more ways to follow your faves? Check out our new player prop tracking with real-time notifications. Or how about more ways to customize your casino page with our new favorite and recently played games tabs? And to top it all off, quick and secure withdrawals. Get more everything with FanDuel Sportsbook and Casino. Gambling problem? Call 1-866-531-2600.

Visit connectsontario.ca. Some news just out on Tesla. The U.S. National Highway Traffic Safety Administration has opened a probe of Tesla's autopilot function. Today Explained, Ramaswaram here with Andrew J. Hawkins, transportation editor at The Verge, which lately means he spends a lot of time thinking about automated cars. Yeah, so we're about 15 years or so into a very broad global experiment into autonomous driving. There's numerous companies that are testing vehicles on the road all across the world.

And interestingly enough, we just got a batch of kind of interesting data from the government about how often these vehicles are involved in crashes. This was data from June of 2021 to May of 2022. What's it say? So they split the data into two sections. There's a section on autonomous vehicles, and then there's a section on driver assist features. Driver assist technology are devices in cars to help the driver avoid a crash.

To sort of sum it up, autonomous vehicles aren't involved in that many crashes because there just aren't that many vehicles on the road. Okay. But the driver assist vehicles are involved in a lot more crashes than the autonomous ones. Like how much is a lot more crashes than the autonomous ones. Like how much is a lot? There were 370 crashes involving vehicles that had advanced driver assist technology. But then for the autonomous vehicles, they recorded around about 130 or so. Got it. Help me out with the taxonomy here a bit, Andrew.

Are the Teslas driving themselves yet? Or are people still driving them? I'm hearing mixed things from the people. You're not wrong to be confused about this. I think Tesla is trying to actively confuse people about what its vehicles are capable of. And that's because of the way that they sort of market themselves. If you buy a car that does not have the hardware necessary for full self-driving, it's like buying a horse. And the only car that has the hardware necessary for full self-driving is a Tesla. What they're actually capable of is what's called level two autonomous driving. Level two? How many levels are there? There are five levels. Whoa!

Level zero. So zero is no automated features at all. Level one. One is the bare minimum. So something like no automated features at all. Level one. One is the bare minimum. So something like cruise control, for example. Level two. Level two is more advanced cruise control. It's things like adaptive cruise control, automatic emergency braking, blind spot detection, lane centering. Advanced features that you find in lots of cars on the road, say. I rent a lot of cars and they don't tell you what your car's level two features

are and sometimes it can get very confusing when the car starts doing stuff that you don't know it can do. It's like a surprise. Yeah it's like a little surprise for you while you're driving a 2,000 pound metal box at 60 60 miles per hour. It's kind of crazy. You'd think there'd be a disclaimer that was like, yeah, your car is going to break by itself, FYI. Yeah, right. Level three. Then you got level three,

which is there are some autonomous features, like the car can technically drive itself, perhaps on the highway, for example. But the driver still needs to stay paying attention to the functions of the driving. It needs to keep their eyes on the road. But for all intents and purposes, the car is technically driving itself.

This is one I hear about, like a buddy got into an Uber recently, and the guy didn't have his hands on the wheel, and he kind of freaked out. Yeah, level three is considered to be pretty dangerous. It's very blurry. There's not, like, a clear line between when the car is driving itself

and when the human needs to be driving it. Level four. And then you got level four. There are level four vehicles on the road today. Google has some in Arizona, for example. These are cars that can operate with no human behind the wheel, but they are constrained, either geographically,

like they're only allowed to drive within a certain geographic area, or they can't drive at night, or they can't drive during certain weather conditions. And then you've got level five. Level five is basically like science fiction and this is a car that can drive anywhere under any conditions night day all types of weather all types of road conditions no matter where it wants to go. These vehicles do not exist today in the world and a lot of people seem to think that they might never exist, that it's a completely fantastical notion.

A lot of the problem that we have with Tesla, because they are abusing the language that we use to describe automated driving technology for the purpose of selling a lot of cars, and it's working for them. But it tends to muddy the water a little bit and lead to normal people sort of being confused about what's autonomous and what's not. Right. Because Tesla is sort of hanging out in the level two to three area, as far as I know, depending on like the laws on the given road that you're driving on. Is that right? That's correct. Yeah. Elon Musk goes out on stage frequently and says Teslas will be fully driverless. They will be level five. About three years from now, we're confident we can make a very compelling $25,000 electric vehicle that's also fully autonomous. He's fully embracing the sort of more outlandish and fantastical ways of describing this technology.

And he has a lot of people who are buying into it. But the truth is that Tesla is level two, just like the Volkswagen that you might rent at the airport that is also basically doing lane correction and telling you when to brake or braking for you or something like that. Yeah, that's correct. It might be a very good level two technology, but that doesn't make it any more advanced or capable than it actually is. The driver still needs to be fully aware and fully paying attention to what's going on. And I think this maybe brings us back to these two reports from the government.

Where did they come from? And what exactly do they say about these features as they relate to the levels? Last year, the National Highway Traffic Safety Administration put out the standing general order to the entire auto industry that they need to start reporting crashes that take place in two sectors of the industry. There's the autonomous sector and there's the advanced driver assist sector. So you've got companies on the one side, autonomous vehicles, and they also want the crashes with the driver assist functions. It sounds like really bad publicity for Tesla. Is it actually doing any damage to the popularity of these vehicles? No, not at all. Tesla is still a remarkably popular brand. They basically sell

every vehicle that they make. You know, and Elon Musk has sort of weaponized the criticism against him and used it as a way to bolster his fan base. People who buy his cars not only are just his customers, but also his supporters. You're Elon Musk. What an honor to meet you. I'm such a fan. They are sort of kind of immune to a lot of this criticism. You can't really say anything bad about Elon Musk that he doesn't then just turn around and turn into some sort of, you know, asset for himself. I know Elon likes to weaponize any criticism against him, but there is a lot of

it out there. A lot of people are rooting for him, but way more seem to be rooting against him at times, especially when you see data like this come out that there have been hundreds of incidents with these, you know, automated driving vehicles. Tonight, Tesla confirming this car was in autopilot mode when it crashed in Northern California, killing the driver. People go, see, there you go. We can't trust the robots to drive our cars. Is that, you know, fear or resistance to self-driving cars rational? It's definitely something that the autonomous vehicle companies are thinking a lot about, because there have been numerous studies and polls that have come out in the intervening years to show that most people don't trust autonomous vehicles. They wouldn't use them. They don't see any use for the technology.

And a lot of it has to do with Elon Musk and the way that his cars tend to crash when they are operating in autopilot mode. But then you've also had the Uber crash. A woman has died after being struck by a self-driving Uber car in Tempe, Arizona. The car was in autonomous mode with a human operator behind the wheel when it hit the woman. I remember this from before the pandemic, maybe 2018. Which is still to date the first and only person that was killed by an actual self-driving car.

Level three. It is going to be a problem. And I think you're seeing a lot of the actual autonomous vehicle companies trying to distance themselves as much as they can from Tesla. They're not using terms like self-driving anymore since he has started to use that as like a brand name for his technology. They want to use things like autonomous or driverless. It is a problem and there is a big communication problem. And a lot of it has to do with how we talk about this technology. But that being said, it's a drop in the bucket compared to the amount of accidents we see from cars driven by human beings. Most certainly. There is 100% a crisis going on today in vehicle safety. The cars themselves are safer than they've ever been

thanks to a lot of this advanced driver assist technology, but also because of the way that they're built. But at the same time, that doesn't make them any less dangerous to the people that are outside the vehicle. And I think we've seen that the size of cars have gotten much bigger over the last decade. People are trending towards SUVs, large SUVs and large pickup trucks, and they're not buying sedans and hatchbacks and crossovers as much anymore. And those vehicles are extremely dangerous to drive. They might be really safe for the person behind the wheel, but for everybody else that's outside, it's much less safe for them. So the crashes taking place with Tesla, the ones that took place with the autonomous

vehicles, it is absolutely just a drop in the bucket, but it helps us understand as this technology improves and as innovation progresses and more and more of these vehicles start to adopt these technologies, are they going to be safer? Are we going to see a potential drop in the number of crashes and deaths that take place because of this technology? That's why we need these numbers so that we can better understand whether this is really all worth it, whether all the billions of dollars that we've spent on developing autonomous vehicles

is worth it in the end, because it really does all come down to saving lives. Andrew J. Hawkins, he's on a whole other level at The Verge. In a minute, we head to the most dangerous road for pedestrians in these United States, and it's got nothing to do with robots. It's Today Explained. Support for Today Explained comes from Ramp. Ramp is the corporate card and spend management software designed to help you save time and put money back in your pocket. Ramp says they give finance teams unprecedented control and insight into company spend. With Ramp, you're able to issue cards to every employee with limits and restrictions and automate expense reporting so you can stop wasting time

at the end of every month. And now you can get $250 when you join Ramp. You can go to ramp.com slash explained, ramp.com slash explained, R-A-M-P.com slash explained ramp dot com slash explained r a m p dot com slash explained cards issued by Sutton Bank member FDIC terms and conditions apply bet MGM authorized gaming partner of the NBA, has your back all season long. From tip-off to the final buzzer,

you're always taken care of with a sportsbook born in Vegas. That's a feeling you can only get with BetMGM. And no matter your team, your favorite player, or your style, there's something every NBA fan will love about BetMGM. Download the app today and discover why BetMGM is your basketball home for the season. Thank you. 19 years of age or older to wager. Ontario only. Please play responsibly. If you have any questions or concerns about your gambling or someone close to you, please contact Connex Ontario at 1-866-531-2600

to speak to an advisor free of charge. BetMGM operates pursuant to an operating agreement with iGaming Ontario. Back at it again with Today Explained. I'm still here. Andrew's gone. Now we got Maren Kogan, senior correspondent at Vox, who just this week published a piece on car crashes involving pedestrians in the United States. There is a pedestrian fatality crisis happening in this country that I think because so many of us are used to driving or used to traffic, we get sort of inured to how bad it's gotten. But the data indicates that we are in the midst

of a crisis. So in 2020, there were almost 39,000 people killed in car crashes overall. It's not good news. The press conference here, they broke down some of those numbers and say that they are going in the wrong direction. We are seeing upward trends in some cases that we haven't seen in 20 years. But the numbers were especially alarming for pedestrian deaths. 6,700 people were killed walking in their communities. And then in 2021, it got even worse.

There were about 7,500 pedestrians killed by drivers, and that was the most in 40 years. Pedestrian deaths have been on the rise for the last decade, but the numbers of deaths really started increasing sharply during the pandemic. According to the Governor's Highway Safety Association, overall crash deaths increased 5% over the last decade, but they climbed 46% for pedestrians. The deaths have been increasing for the last decade. We can talk a little bit about why that is, but those numbers really sort of exploded during the pandemic. Well, let's do the last decade and then the pandemic.

So what's been happening in the past 10 years? The biggest problem is that America has continued to structure our life around cars and not people. If you look at most of the communities where people live, they're built for drivers and not pedestrians, outside of a number of rare small towns and cities where things are different. So I think what would be really interesting is for people who are listening to look in their community and keep an eye out for something called a strode. A strode? A strode. A street-road hybrid. Oh. There's this guy, Chuck Marone. He was a traffic engineer. And he describes roads and streets as two different things. So if you think of a road, a road is meant to get people

from point A to point B as quickly as possible, right? So it's meant to move people from one place to another. So a road that's built for cars should be high speed. It should be efficient. It should be clear and easy for drivers to get from one place to another. A street is something different, right? A street is a place. If you think of like a nice quaint downtown, you think of a street where people are able to go out to eat, they're able to play, they're able to shop. It's a place where you want to have the traffic be very slow so that it's safe for people on foot to get around. And so that the traffic coming through is forced to slow down and sort of commune a little bit with the businesses and the people and the life in the community.

What we have in America is a bunch of stroads. Stroads are multi-lane. They have lots of traffic coming through, so they're always gridlocked. There's a lot of stop-and-go traffic, so cars are neither moving efficiently or safely because the other thing is that they often have a lot of commercial development on either side of them and a lot of people walking around to get to those commercial developments. So what happens is you create all of these opportunities for pedestrians and cars to come into conflict with one another. And we see these in every community in America.

Strodes are very, very deadly for drivers and passengers and cars, but especially for people who are interacting with them outside of cars. And this probably brings us to pedestrian deaths going up in the pandemic. It does. I think one of the big things that happened during the pandemic is that the usual traffic patterns were disrupted. So there's a lot of congestion on these stroads,

right? A lot of stop and go traffic. We think of congestion as normally a bad thing, but it's actually a really good traffic calming measure. It's really hard to drive super fast and kill someone if there's a car directly in front of you, right? That slows traffic down. So what happened in the pandemic is all of these usual traffic patterns were disrupted. It opened up the roads and allowed people to drive a lot faster on those roads. And at the same time, you had a lot of people getting out walking and biking, right? Like I became a runner during the pandemic. I started biking a lot during the pandemic. So you have this sort of perfect storm for pedestrian risk. And this is what you write about in your most recent piece titled The Deadliest Road in America, which is about this one stretch of road that's super deadly

for pedestrians. Tell me about it. It is a stretch of US-19, which runs north from Pennsylvania all the way down to Florida. This is one part in Pasco County, Florida. U.S. 19 has gained a reputation over the years and a nickname, Death Valley. I refer to it as Murder Highway. A study was done last year. They took all of the government's data that they've collected on pedestrian fatalities and tried to identify 1,000-meter stretches of road where six or more people had been killed over a 16-year period. When they identified all the hotspots,

seven of them were on the single road in this single county. I mean, they were shocked by that result, and they looked at the numbers, and they were like, you know, 137 people have died in these hotspots. That's like a commercial airline going down. And, you know, that would be national news. Did you visit this road or is it too scary to visit? I visited the road. I drove the road. I walked the road. I interviewed people who have lost loved ones on the road. Yikes. Yeah, it was, I will say it was much more pleasant to drive the

road than to walk. I sort of was really hating my decision to walk the road when I did. And what is it about this particular road that makes it so deadly? It's so many lanes. There are three lanes going in either direction on either side. And then there are also multiple turn lanes on either side. So if you get to an intersection, there are nine lanes, and a pedestrian has to cross this incredibly huge expanse. The other problem is that the crosswalks at the lights are super far apart. So let's say you're standing on one side of the road, and you just want to go to the Publix across the street. You can either walk a mile to get to a safe crosswalk and back, or you can make a run for it. Now, at the same time, there are all these lanes and it's very

wide, flat, and open. The speed limit is as high as 55 miles an hour, which means a lot of people are going 60 to 70 miles an hour. So everyone is in a way sort of acting rationally. The drivers are driving fast because the road is giving them cues that say you can drive fast. The pedestrians are running across the street because they're saying, it makes no sense for me to walk all the way to a crosswalk. So you've created this really dangerous condition

where people are running across the street and drivers are going too fast to be able to stop and see them. Does this one stretch of US-19 tell us about car infrastructure in America, or is it sort of a Florida thing? Well, it's both a Florida thing and a story of America, I think, in the way that so many Florida things are. I talked to one traffic engineer, and he said, you know, you can think of Florida as, you know, on the leading edge of so many different trends. And because there's so much growth and so much development, we see a lot of things before the rest of the country does. But, you know, in reality, there are stroads like US 19

all over the country. So for all the people who just don't want to accept this American normality of traffic fatality, like what can they do? One thing is improving public transit, making it easier for people to take public transit because public transit is safer. Investing in bike lanes and allowing people to travel by bike because people on bikes are less likely to kill people than people in cars. You know, other countries use traffic cameras to great effect. Car manufacturers could do more. They could add speed governors to cars. They could force cars to slow down

and, you know, employ a number of street design features like traffic circles, curb extensions, speed humps to help calm traffic and encourage drivers to slow down. One last thing I want to ask you about, and it's something I actually heard about from you in a conversation on the Vox Conversations program with the journalist Jesse Singer. You two talked about the word accident as in car accident. When we say it was an accident, we're not saying it was random. We're saying it wasn't my fault.

It wasn't their fault. And in doing that, we're almost always focusing on the wrong thing and setting up the same accident to happen again. Because fault and blame and even absolution has nothing to do with the problem of accidental death and injury. Accidental death is a matter of dangerous conditions.

So if you're talking about bad people and good people, who's wrong and who's right, you're barking up the wrong tree. So what should we call them instead? Crashes. Crashes. Crashes. Interestingly, when cars were first sort of driving on streets and someone would say, hit and kill a child, they didn't call them car accidents. They called them car murders. And they called them child car murders. And mobs would surround the vehicles and try to pull the driver. The drivers would have to be rescued by law enforcement because people were

so outraged. And then the car industry introduced a term called jaywalking to shift the blame onto people who were walking and make it seem like it was their fault. And so psychologically, there's something very tempting about being like, well, it was just because that person was being an idiot and not because we have these dangerous conditions. So let's dispense of accidents and also jaywalking. I'll just be like, I'm walking. This isn't jaywalking. It's just walking like, I'm walking. This isn't jaywalking. It's just walking. You're just walking. I love it.

Be careful. Marin, Kogan, Vox. Earlier, you heard from Andrew J. Hawkins, Verge.

Our show today was made by John Ahrens with help from Matthew Collette, Laura Bullard, and Paul Mounsey. This is Today Explained. Thanks for watching!