Tech Isn’t Fixing the Crisis on Our Streets
Paris Marx is joined by David Zipper to discuss how Silicon Valley pitched new technologies as the fix for a whole range of transport problems, and how that really just distracted us from solutions while allowing issues like road deaths, emissions, and traffic to get even worse.
David Zipper is a Visiting Fellow at the Harvard Kennedy School's Taubman Center for State and Local Government and a contributing writer at Bloomberg CityLab. You can find his articles and sign up for his newsletter at DavidZipper.com and follow him on Twitter at @DavidZipper.
Support the show
Venture capitalists aren’t funding critical analysis of the tech industry — that’s why the show relies on listener support.
Become a supporter on Patreon to ensure the show can keep promoting critical tech perspectives. That will also get you access to the Discord chat, a shoutout on the show, some stickers, and more!
Paris Marx: David, welcome to Tech Won’t Save Us.
David Zipper: It’s a pleasure to be here. Thanks for having me.
PM: Very much looking forward to talking with you. Obviously, the listeners will know that transportation topics, transportation and technology come up quite a bit on the podcast because it’s something that I am interested in. And I’ve been reading your work on these topics for quite a while now. So I’m very happy to speak with you today. And I’m hoping that to get us started, you can give us an idea of the problems that really exist on the roads today before we start talking about these tech solutions. What are these issues that people are often responding to even in the tech industry, when they recognize that something needs to change about the transport system, even if their solutions aren’t always the best way to do it? What are the problems that are out there today?
DZ: The problems with with road transportation? Well, I think there are many [laughs]. We could spend a long time talking about them. But I guess the buckets that I generally refer to would be climate. Road transportation is the biggest source of emissions in the United States. And I would not be surprised if you told me it was the same case in Canada where I know you are at the moment.
PM: Number two in Canada — second to oil and gas because of the Oil Sands.
DZ: That would make sense too. Climate is a huge problem with our road transportation network. So is safety — this is one area where actually the US is in a different trajectory than Canada. I wrote an article about this in CityLab. The US just hit a 16-year high in annual traffic deaths in 2021, up to around 43,000 in a year. And what’s really driving that growth is pedestrian and cyclist deaths. Cyclist deaths rose 44%, in a decade in the US and pedestrian deaths hit a 40-year high last year, which is just incredible in the United States. And in countries like Canada and the European Union, things are not that bad. Usually traffic deaths have been going down. And almost all these countries have been going down in the last 20 or 30 years. Unlike in the United States, my country is uniquely terrible at road safety. Before some of your non-American listeners pat themselves on the back, be careful, because some of the bad habits of American road transportation and safety are spreading. So we can talk maybe a little bit about that later.
And then I would also reference equity, which is a little more nuanced, related to road transportation. But when you design cities and design a society such that you really can only access a good job if you have a car, you’re making it very difficult for those who have low-incomes, who are often minorities to be able to improve their wages. I think of a really interesting study that was done by a guy named David King, a professor at Arizona State, who showed that if you’re low-income, and you don’t have a car in New York City, you don’t really pay a wage penalty in terms of your ability to get a better job, because in New York City, there’s pretty good transit. However, if you’re in Phoenix, and you don’t have a car — major penalty. You’re going to have a really tough time getting a better paying job. And unfortunately, in the United States, again, which is where I live, and where a lot of my work focuses on, we have a lot more Phoenix’s than New York’s, if you will, which is a huge problem in road transportation. So those are some of the issues I think about as being underlying our road safety world.
PM: I think you’ve laid it out really well. And we could also look to traffic, the amount of time that people spend stuck in traffic, the health consequences that come with being stuck in cars all the time. There are a whole load of things that we can discuss. I think your point on equity is really important, because I feel like it’s a part of the conversation that doesn’t get looked at enough. We often don’t consider what the consequences of building these cities that are so dependent on cars are on people’s ability to access the services, the jobs that they need to access. And because auto transportation, auto ownership is so normalized, and so many people own one, it’s often hard to remember or to think past, the difficulties that presents the people that people who can’t afford to own a car, what it means for them. And even the people who own a car, because they feel like they have no other choice. But then it’s a massive expense for them, and leaves them in a more precarious situation. Because they might have this car that doesn’t work very well, that’s old, that needs frequent maintenance. And so then there’s always the risk that something’s going to happen to it. And you’re living paycheck to paycheck, and it’s just going to put you further behind, maybe you won’t be able to fix it. It’s a serious problem.
DZ: It really is. And it’s something that I think is a little hard to wrap your mind around, if like probably a lot of listeners, you don’t have to worry about being able to afford a car. Most of us just can, who are going to be able to listen to a podcast like this. But when you really can’t or you’re really struggling to do so maybe one breakdown can set you into financial tailspin where you lose your job and you don’t have the money, you need to fix the car and now you’re unemployed. And you can end up in really precarious positions. And meanwhile, cars are just getting more expensive. That’s the other point too. And the United States, monthly car payments just hit, the average one hit $700 US/month, over the summer an all time record, which is a combination of automakers focusing on SUVs and trucks which are more expensive and less so on sedans, which are less.
In fact, companies like Ford don’t make any sedans or offer any sedans in the US at all anymore. And meanwhile, interest rates are rising after being low for very many years. So I actually wonder if there’s a bit of a reckoning coming about the affordability of cars. The problem is, in so much of at least North America, there’s not really a great alternative. After so many years of designing in car-centric ways and providing such paltry transit access and bike lanes, I really worry what’s going to happen if cars become much more expensive. And we still have generally failed in places like Dallas or Cleveland or what have you, to be able to really provide strong access to jobs or just daily necessity trips.
PM: It’s so important. Because I always get the Canadian and the American numbers mixed up here. But the AAA and CAA in Canada, they often estimate the annual cost of car ownership and it’s something like the high single $1,000/year or even over $10,000, and into that range depending on the type of car that you own, whether you own a sedan or a truck or an SUV or something like that. But that’s a real huge expense for people, especially if you’re thinking about people making minimum wage or a little bit over minimum wage. That can be as much as rent in some cases. And of course, rents have soared as well. So it just shows how difficult it is for a lot of people. And you mentioned before as well, the safety piece of this. And you said that within North America, Europe, these sorts of countries, the United States is really an outlier in terms of safety and road deaths. And the number of people who really get hurt on the road, why is there such a huge difference in the United States versus other countries? And why is the United States going backwards instead of improving these these things?
DZ: It’s a really important question. And it’s one that I wish more Americans asked, I don’t even think most Americans who work in traffic safety are aware of just how bad the United States is when it comes to road safety. I’ve done a series in the last few months with CityLab, looking at traffic safety outcomes in other countries, doing articles about France, and Finland, Japan, and Canada, your country. And the differences are stark, basically European countries in Japan are between three and four times safer on a per capita basis when traveling than in the US. In other words, US per capita death rates and crashes are three to four times higher than in those countries. And then people say: Oh, well, the US is car-focused, and we’re spacious, so it’s not fair to compare the US to Japan. Okay, well compare it to Canada. Your country is pretty spacious, and car-focused, last I checked, and still, the average American is two and a half times as likely to die in a crash as the average Canadian. So your question is a fair one. Why is the US so bad at this? And I’ve been trying to really wrap my mind around it with this series. By the time this podcast comes out, I will have published a culmination looking at some overall lessons.
And also I should emphasize too — the US is trending in the wrong direction, while the rest of the world, the developed world, is trending in the right direction. Americans are getting at an increasing risk of dying in a crash while those and other countries are getting safer. So what’s really changed? Well, there’s a few things that other countries have adopted that the US has not. Some of it has to do with kind of just improvements in infrastructure, like roundabouts, which have had a huge impact on road safety in countries like France. And the US, outside of a few enclaves like Carmel, Indiana, they’re pretty rare. You can look at car-free streets and low-speed, urban streets, which are much more popular in Europe or in Japan and Canada in the last 20 years than they are in the United States. And then I think something that’s a little bit hidden — you have to actually understand the trends to see the impact it can have — is the relative trends in train and public transportation ridership, because and I wrote about this as well, people don’t necessarily think about it, but you’re actually between 20 and 60 times safer taking a bus or a train on a given trip, than you are driving. So robust transit systems, and a lot of people taking transit or riding trains, means you’re going to have a lot fewer opportunities to crash. So when you look at the trends, it’s pretty striking.
Even before the pandemic, the US, unlike Europe, and I believe, unlike Canada as well, saw a decline in transit trips, a mode shift away from transit and toward driving — all else being equal — that’s going to increase the risk of dying in a crash. And then you look at a country like Japan, where train ridership is through the roof, because there’s just such great networks. That’s been a huge driver in the drop-off in crashes there over the last 40 years. Japan’s now at an all-time low and fatalities. And then the last point I’ll make and I have a feeling you’ll want to talk a bit more about this, in terms of why the US has seen an uptick in crash fatalities that exceeds other countries. My country is obsessed increasingly with SUVs and trucks that are themselves growing bigger and bigger. I talked to a Canadian who says: Well, we have them too, but they’re generally a size smaller. And in Europe, they’re a lot smaller, although they’re growing and they’re becoming more popular.
And the problem with large SUVs and trucks, and this is well documented, they’re terrible for the safety of anybody who’s not inside one — particularly pedestrians and cyclists because of the blind spots of these vehicles, both the A-Pillar which is basically the piece of the frame between the windshield and the side door window. And because they’re getting taller, you can’t see people who might be in front, particularly a child or in a wheelchair. And because they’re also getting heavier which creates more face in a collision. And their height — their growing height means that the front end of the vehicle is more likely to strike a person’s chest, their torso, which can be very deadly, as opposed to hitting their legs, which a sedan is more likely to do, which is not good. It’s not fun. But you’re more likely to survive that if you’re hitting your legs rather than in your chest or your head. So those are some of the reasons why I think the US has seen a particularly bad performance in road safety over the last 20 or 30 years.
PM: It’s always surprising to me when I hear people say that they feel more comfortable or like safer driving a truck or an SUV or something like that. I can get driving a truck for work purposes, you need to tow things and whatever. But the times that I’ve driven a big SUV, I felt terrified because I can’t see everything around me that I feel like I need to be able to see in order to make sure that I’m not going to hit another vehicle or a person or something like that. Personally I find it terrifying. But I guess maybe you get used to it. I want to drill down on this point about the US and Canada just for a second. Because I certainly don’t want to make it seem like Canada is some traffic utopia or something like that — because it’s far from it, right?
DZ: Americans do tend to fetishize Canada, sometimes. Americans of a certain ilk — you have no guns, everyone’s healthy. You’re also friendly and nice. That’s not true, really? [both laugh].
PM: Thank you very much, no — not totally. But people who do activism around transportation issues have certainly identified a ton of problems in Canadian cities. We’re still very reliant on cars; there’s still a lot of suburbia; our intercity rail system is terrible up here, like in the United States. And I believe transit ridership is a bit higher in major Canadian cities, but I don’t know if it’s so much higher than the United States. And when I would think of a Canadian city versus an American city, okay, we have some bilingual signage and stuff like that, and whatnot. But, in terms of what’s materially different, I can’t think of things that are so different. So why would Canada be so much less or there would be so much less of a risk of dying in a crash than in the United States?
DZ: I’ll point to a couple things. And again, there’s more information in this article that I wrote, I think it was in July, looking at Canada versus US. One thing is that the price of gasoline is higher in Canada because of tax policies. And that has contributed toward people wanting to drive less. The average American drives a lot more. More driving means more chances to crash. This is my hypothesis. Perhaps some Canadians will educate me on this. But my sense is that the relatively high price of gasoline has contributed toward Canadian cities being a bit denser, which creates a virtuous cycle with transit service. When you have a denser city that’s bigger, you get better transit service, you make it more easy to travel without relying on a car. And I can also say that, because I looked at the data here. Canada is a more mega-urbanized country than the United States, if you will. The percent of the population that resides in the three megalopolises, if you will — Toronto, Montreal and Vancouver — is significantly higher than the percent of Americans who reside in the megalopolis is of New York, LA and Chicago, even if you go further down. And Canada is actually more urbanized, despite all of the rural spaces that it has.
The other point that I found really interesting. And to be honest, I didn’t know about this. It has to do less with the levels of fatalities in Canada in the US, but the trends over the last 20 years. And something I will give Canada credit for which the US has just not done. And this is not as I understand as much national policy as provincial, but a lot of the provinces in the last 15 years shifted toward immediate and pretty serious punishments for people who are caught recklessly driving or driving under the influence — your car is impounded immediately. And the penalties have gotten stiffer and more automatic. And that has had a big impact, from what I understand, on people’s willingness or acceptance of driving drunk — which is like a third of traffic crashes in the US, by the way — or recklessly driving. In the US, these sorts of penalties are going to vary from place to place. It could be quite a while before that penalty is going to be instituted, and you could get a lenient judge. And I think the variability is a lot less. And the punishment is a lot faster in Canada, which leads to more of a deterrent effect. And that’s a trend in the last 15 years. and Canada as I understand it that’s been quite impactful that we just haven’t seen in the US in the same way.
PM: That’s fascinating. And I feel like a lot of those points are accurate, line up with things that I’ve observed. Certainly Canadian cities have a lot of suburbanization. There’s still a lot of sprawl in our cities, but there is at least a dense core and I feel like a bit more density going out from that as well. But we can certainly discuss this aspect of things for ages. But I want to move on to talking about the technology. Because I feel like in particular, over the past decade or so, you know, one of the big narratives that we heard was that autonomous vehicles, were going to arrive in just a couple of years if we would give tech companies a few years to perfect the technology. And then we wouldn’t have to worry about these problems with the roads or particular political decisions around transportation, because they would address traffic. They would address road deaths. They would address the inequities in the cities, as you were talking about earlier. And all we needed to do was wait for this magical technology to arrive. And that never really worked out as planned. How do you look back at that kind of period of time when this was the big promise that was being made from the tech industry and what actually came of it?
DZ: It reminds me — somebody wrote something about Hyperloop getting in the way of high speed rail. That’s what I’m thinking of! [both laugh]. For those listeners out there, Paris wrote about this, you should go read his article about it. But I wrote a long piece about what’s the purpose of self driving cars in the Washington Post, in February of this year. In that article, I went into through the history of autonomous vehicles, which was very much not based on solving the problem of road safety or anything else. The original rationale or impetus for developing autonomous vehicle technology was the cool factor. It’d be really neat if we could make these vehicles move on their own.
PM: Looks like science fiction!
DZ: Wouldn’t that be great! That’s what people at Carnegie Mellon were thinking and, it’s developed from there. And they have a famous Robotics Lab at Carnegie Mellon. And then, the military in the US said: Actually these could be really useful in battle environments. So the DARPA Challenge happened that was funded by the military. And then Google got involved in basically snapped up some of the people who were central to the DARPA Challenge. And it was only once Google was involved that you started hearing this narrative, that autonomous vehicles will solve the road safety problem and save the 100 people a day who die in road crashes in the US. Let’s be really clear. This was an after-the-fact public explanation of why Google was doing this, which, frankly, when you unpack it is quite hollow. Because there’s really no way to know whether or not truly self driving cars will be safer or not than human drivers. Not only because people are actually quite talented at being able to handle unexpected things, which computer systems are often not, but also computers make mistakes that humans never would. The best example I was given was a computer system that kept not noticing construction workers, and they were trying to figure out what is wrong with this system? What is going on? It turned out that the system just wasn’t observing the color yellow, which the people were wearing the construction workers. You’d never know that because the machine learning doesn’t actually tell you what it’s missing, you just can observe that it’s screwing up.
So the safety is not a great argument, or a compelling argument to me for self-driving cars. And then there’s others that you hear around equity, or particularly around the environment is truly nuts, because the one thing we really do know about what self-driving cars are likely to do is that they’re going to induce a lot more driving. Because when things get easier or cheaper, we want more of it. We find new ways to consume it. And there’s already been studies about this showing that with self-driving cars, it’s just like, if you had a chauffeur you’d take more trips than you would without one. And when you imagine a world of self-driving cars, where people own them, that is a terrifying world for cities, because it’s going to be gridlock like you’ve never seen before. And I don’t care if these cars are electric or not. If you’re going to see a huge uptick in driving, that’s going to be really bad for the planet. So I am quite dubious about whether self-driving cars are actually going to solve problems or if they are a solution in search of a problem.
PM: I think you’ve put it really well. And I think another piece when you think about the energy requirements of it — and I feel like this is a piece that doesn’t enter the narrative, because it’s not part of the PR story that the companies wanted to tell us — was really the energy requirements of having this really high power computer in every car, having to drive it around. And then also how they’re collecting just wild amounts of data, and sending that back to data centers where it all has to be stored and processed. And so there’s all of this kind of impact that goes along with it, as well. And certainly that’s being built into cars today, too.
DZ: I think that’s all true. You could keep going to all the various problems raised by AVs. But what I take a step back and think about it, is to say: Well, why don’t instead of starting with the technology solution, which seems to always, “solution” by the way, that orientation always seems to get us in trouble when you start saying: How can we use Hyperloop? How can we use self driving cars? How can we use flying taxis? Instead, why don’t we say what are we trying to achieve societally? And what’s the best way to go about doing that? It’s like counterposing FOMO with, the term I use is, mundane mobility. FOMO is how you end up with city officials and state officials, and even countries falling over themselves to get Hyperloop or autonomous vehicles before their neighboring jurisdiction can. And mundane mobility is the stuff that doesn’t grab the headlines. But it’s so necessary for achieving goals around climate, climate support, or enhancing equity, or reducing road deaths. It’s things like expanding sidewalk networks, or expanding bus frequency or building bus shelters. This is basic stuff that is not going to make you if you’re a mayor, or a governor get a front page story in The New York Times, but it’s a heck of a lot more likely to actually improve the lives of your own residents, and, frankly, improve the planet.
PM: I completely agree with you. And I think it’s frustrating that that’s the way it plays out. It’s a lot better — it’s a lot more attractive to be trying to attract the Hyperloop for the Boring Company, or what have you, because it plays well, because you’ll get a lot of media attention for it. Because there might be some voters who would look at that and say: Okay, that’s innovative, this person who’s trying to move things forward, rather. But then if you take these kinds of steps that are actually going to solve the problems that are mundane, that are not playing on whatever someone in Silicon Valley is saying is the next big thing, that might actually solve some real problems in the city that might make transportation better, that might save some lives. But it’s not going to be sexy. It’s not going to get the headlines in the same way or the same kind of attention. And it seems like a real problem with the incentives that are built in how we think about transportation and how public officials approach it.
DZ: I think that’s right. And so you say, what do we do with that is the next question. And if you’re listening to this podcast, you probably would like to see your elected officials focusing on the mundane, as opposed to the flashy. And a simple thing you can do is to help build a constituency of people who celebrate the mundane. Maybe there’s a way to actually build a big sidewalk as as a lobbying group, I joke but we frankly really need that. When you look at how essential sidewalks are to infrastructure networks, wouldn’t it be great if somebody who’s running for office is like: You know what? I really need to be able to get support from the people who are thinking about making it easier to walk. I don’t think it’s happening now. But there’s no reason why you can’t. And I think that grassroots organizing, even tweeting support for a city councilor or a mayor who does the right thing and says: Actually, we should think about reallocating some of this money toward a rail expansion, toward improving sidewalk networks in our low-income community. That’s something that needs to be called out and supported. So that people who do take those positions that are kind of brave right now, realize they’re gonna get some credit for it.
PM: It’s always cool to see those occasional stories of people painting on their own crosswalks in places where they think that they should belong, or expanding the sidewalks in their areas by putting down cones or kind of painting something on there. So it looks like a car shouldn’t go there to show that see, things still work perfectly fine like this, but it’s better for the people who use these roads in the streets and things as well.
DZ: Have you had Peter Norton, the historian, on the podcast before?
PM: I haven’t, I should though.
DZ: You probably should. I learned a lot from him. And for those who don’t know it, his book “Fighting Traffic” is a bit of a Bible for a lot of people who think about why cars dominate cities. It’s a wonderful history, looking at how what he calls “motordom” took over city streets in the 1910s, 1920s, and relegated pedestrians and children to playgrounds, sidewalks and crosswalks. It’s really fascinating hearing from him, as he describes how there were these mass movements in the late 1910s and early 1920s, of like hundreds of 1000s of people in individual cities who would do these marches and erect monuments to traffic victims or crash victims, and they’re all advocating for pedestrian rights. This is how you actually build power and motordom was successful, and he explains how, in undermining that and establishing car speed as a public goal in cities, which I think is sort of the original sin of the United States in our urban design, at least with our streets.
But the point becomes, when you think about the potential to really galvanize the masses to push for change, that’s what we really need. Peter talks about ghost bikes being a step in the right direction where you paint a bike white at a place where a cyclist was killed. And the whole point is to show this is a public loss. This is not just somebody who lost their life, that’s the end of it. It could have been any of us, and this is a societal problem. So I strongly support basically the grassroots movements around these topics, because I frankly, think without them, I don’t know how we’re really going to make progress.
PM: And I think that there’s a lot that we can learn from those histories, whether it’s the 1910s and 20s, or the 1970s. And these periods when there was really kinds of backlash to the domination of the automobiles, or the growing domination, and trying to restrict that, and how we can learn from it today. I want to move on to talking about some of these other technological approaches, because I feel like we see the failure of the autonomous vehicle, really — sure, they’re still out there being tested in some cities, and what have you — but they’re not going to revolutionize transportation in the way that we were sold a decade ago. And I feel like advanced driver-assistance systems (ADAS) really come out of that, right. So what are they? And how do they build on it? And what is the promise for what they’re going to mean for transportation?
DZ: So ADAS is is the lower hanging fruit compared to autonomous vehicles. And ADAS is becoming more important. Because as we’re recording this, it was pretty recently that there was a big shock wave for the AV industry, when Argo AI, an autonomous vehicle company, which had billions of dollars in valuation just closed up shop suddenly. It was kind of stunning. And in doing so the Ford CEO said: We’re going to focus more on ADAS than we are on fully autonomous systems, because we think that’s more feasible and more marketable. And so what ADAS is — maybe it’s helpful to explain that first to the listeners. It’s sort of a collection of features, some of which will be pretty familiar to folks like adaptive cruise control, and Lane Keep Assist, automatic emergency braking, pedestrian detection. And these individual features, some of them are meant to prevent crashes, some of them are meant to just be conveniences, really. But when you knit them together, in a powerful way, you can end up with what are known as level two ADAS systems level two according the SAE definition of autonomy. And these systems are things like Tesla Autopilot, or Cadillac Supercruise, which can let the car be driven and operated on its own, as long as the driver is always vigilant and always ready to jump in, when something wrong might happen.
And these systems have what’s called a driver monitoring system to keep track of whether that driver is paying attention, either by looking at the angle of their face and where their eyes are, or how they’re gripping the steering wheel. So that’s what ADAS is. And it has been under development, really, for decades. You can go back to cruise control. Cruise control is sort of a form of ADAS, if you think about it, and that was run in the 80s. Now it’s a much bigger deal. And I wrote an article for The Verge in September about how car companies hold up ADAS as their solution for the American road safety crisis. This is how we can end up with zero collisions, which is part of the vision of GM. That’s their defining mission at the moment. And I’m pretty skeptical about it for a variety of reasons which we could go into. But I’ll let you decide if you want to delve into that. Now, if you had another question to tee up first [laughs].
PM: No, I think it’s good to dig into it. Because part of this is the lane assist systems, the pedestrian detection — the idea is that the car will be able to detect all these things that are going on around it. And really help the driver to navigate these things even without having to control the vehicle unnecessarily or to give them kind of nudges and alerts that things are happening. It’s interesting, I’ll tell a quick story I actually used to work in car insurance. This was a few years ago now. But I remember, I had a woman call in one day, and she was just completely shocked. She had just been in an accident, or crash. And what had happened, as she described it to me was that she was in the vehicle. There was a vehicle up ahead. And the car had a system where it was supposed to detect the car in front and slow itself down because it was on the highway. But for whatever reason, in this instance, it didn’t do that. And it caused her just to ram right into the back of the other car, which was stopped and stuck in traffic. She was obviously affected by the fact that she had been in a crash and this had taken place. But even beyond that, she was like: The car was supposed to stop; I don’t know why it didn’t stop like I was expecting it to. So it sets people up to have these dependencies on the system that it can’t always follow through on. That’s just a story but feel free to give us a bit more info on this.
DZ: It’s an illustrative example because it hits on some of the real problems with ADAS. First of all, some of these systems do not work reliably. AAA did a study where they found that pedestrian detection is basically worthless at night, which is when the majority of pedestrians are struck and hit the cameras and sensors don’t work effectively enough, it also can’t work. Because of physics, a automatic emergency braking and pedestrian detection can’t work above like 35 miles per hour. So that’s a limitation. But then you factor in the fact that you could potentially improve these systems to a degree. But still, people don’t always understand what the system can and cannot do. So the woman who called you may have overestimated or misunderstood what her system could do, this happens all the time. Or maybe she didn’t, the system expected her to take over because remember, you always as the driver responsible with a level two system, you have to be able to take over.
And there’s a lot of evidence accumulating now, that ADAS systems lead driver skills to atrophy from lack of use. Driver skills in emergency situations degrade. If we’re all becoming worse drivers, because in an average minute, we’re doing less on the road. That’s really bad for the occasional emergency unexpected situations that pop up. Those are a couple of the big concerns. But I actually think there’s even more than that. There is a well known economics theory called the Peltzman effect, it goes back to the 1970s that shows that behavior compensates for mandatory safety features like seatbelts, if you have a seatbelt in your car, you’re more likely to take risks as driving because you assume that feature is likely to to protect you, that might be true for you. But it certainly puts other people who are not in a seatbelt and car at risk, especially pedestrians, and cyclists.
So that’s something that makes me worried about ADAS and that way to where it might lead people to drive more recklessly in ways where ADAS can’t fully compensate for the risks that they’re creating. Especially by the way, if carmakers are leading people to misunderstand what their AI system can do, and for anyone who’s following how Tesla advertises autopilot, and quote unquote, “full self-driving,” they sure do seem to be creating some misunderstandings about what these systems can and cannot do. And then the final point about why I worry about ABS is, I think, actually intuitive when you take a step back. And that’s that a lot of these systems like Lane Keep Assist, or adaptive cruise control really aren’t related to safety. They’re conveniences that make driving more pleasant or less tedious. And when things become less tedious, we just we do more of it. And if ADAS leads people to just drive more miles, kind of like we’re talking with autonomous vehicles a few minutes ago, that in of itself is a recipe for more crashes, more miles driven — all else being equal — is more chances to hit something or someone. So all of this is to say ADAS is not a panacea for road safety. It’s not even clear how much a perfectly functioning AI system is going to reduce crashes, eliminate some crashes.
On net, this is not to me, the solution for the US road safety woes, particularly, when you, again, take a step back and say: Forget about ADAS as a solution. Let’s start on what we’re trying to achieve, which is a reduction in road deaths. If we’re starting with what is going to reduce road deaths, then you can say what’s the most effective way of doing it and you get into things like speed cameras, and speed limiters. And you look at bike lanes, and you look at eliminating cars from dense urban areas. These are not the things that car companies want to promote, because it doesn’t enhance the value of their core product. But these are far more cost effective ways of actually reducing the maximum number of people whose lives are lost and crashes every day.
PM: When you were talking about people taking riskier actions, because of feeling that they’re safer in the car with things like seatbelts. What I immediately thought of was autopilot and the videos people — who are supporters of Elon Musk, who are listening to him — are putting out where they’re clearly doing very dangerous and reckless things, just to try to prove that the system can do more than it actually can. Because they’ll find these kind of perfect situations where it might work perfectly using it on the highway or something like that. But where there are real risks, not only to themselves, but the people around them. You bring up a Tesla and autopilot. Another thing that they have really kind of pushed and pioneered and led other automakers to adopt is this move toward the infotainment system that is on the touchscreen, where you’re taking away the buttons that are in the car. And so more and more of what you need to do the controls for the car are just on this touchscreen. There’s no kind of physical feedback for you. And it seems like more automakers are moving in this direction to try to replicate this thing that Tesla has done first. There’s even a recent Apple demo showed changing the whole dashboard into a screen that you could customize, and the promise is that this would reduce distraction because people wouldn’t have to look at their phones anymore. But it seems like it’s doing the complete opposite.
DZ: Yup! You must have looked over the list of articles that I’ve written because this is one that I delved into deeply. By the way, I’ll just mention I’m referencing a bunch of articles as we chat here. If anyone’s curious, I’m sure Paris will put them in the notes. But they’re also all on my website, which is davidzipper.com. But I wrote, I guess it was a year ago now, a deep dive, I think the first deep dive into the safety risks of infotainment systems in Slate. And you’re totally right. This was a choice that car companies made that people didn’t even really ask for. But car companies decided that we have to compete with Apple and the iPhone. So we’re going to create touchscreens in place of the haptic feedback and tactile nature of the knobs and dials that I grew up with, when and how I first learned how to drive way back when. And this is just flat out a step backwards for safety. Because if you can change the song that’s being played, or make adjustments to air conditioning without taking your eyes off the road, because you have to look at that touchscreen that is a second or two, with your eyes off the road that is absolutely creating more risk of crashing. And there’s this issue of will people be on their cell phones otherwise? Okay, well, then that’s a problem that we should probably be talking to Apple and Samsung and others about and erecting regulations.
But as it is now, I don’t know what it’s like in Canada, but in the US, NHTSA, which is really the authority that that can regulate this has been pathetic in its response. There is voluntary guidance that was issued a few years ago that set maximum thresholds for the amount of time that one’s eyes should be away from the road for every step of an overall task and how long it takes to complete that overall task. And immediately, carmakers violated that guidance without penalty, because they want to have a distinctive feel within their cars. I don’t think car companies are trying to kill anyone, actively. But they’re striving, particularly in this age of electrification, to find ways to differentiate their core products. And when you take away the roar of the engine and the feel of that gasoline powered system in your car, the feel of sitting in the front seat, and what that dashboard infotainment system looks like is a key differentiator. And unfortunately, that’s sort of unleashing an arms race that is leading to some really weird stuff.
There was one system that everyone seems to hate — this is BMW’s word — it’s like a gesture identification. You can wave your hands to do things. It doesn’t work right. I have someone who has it tell me he’s convinced his car is possessed by a demon. And he went to the dealership to see how they could get it removed. But this is all an effort to differentiate. And I think it’s worth taking a step back and saying, from a societal perspective, is this really where we want competition? Or should we be putting in some constraints to ensure that competition may be improving the performance of cars and making them more fuel efficient, making them better in certain ways. But I’m not sure that variegated infotainment systems is something that is going to be societally constructive.
PM: Hearing what you’re saying there, what immediately comes to mind is it feels like these automakers and this industry really needs to be reined in. It makes me think of, say, Ralph Nader in the 60s and 70s. And how that really kind of gave a punch to the industry, made them wake up, made the regulators and the government take safety seriously, again, or for the first time maybe. And it feels like there’s been this kind of drifting for a while. And we’re at this point where, especially with electrification, with the advent of these new technological systems that are being put in cars, it feels like there really needs to be like a wake up again, to be like: Look, we really need to pay attention to what is going on here, to what the potential effects of this are, especially when, as you say, in the United States, things are going in the wrong direction, not the right one.
DZ: That’s right. And by the way, it’s coming for you too in Canada, and it’s coming for people in Europe, where cars are shifting more toward SUVs. They are getting heavier, infotainment is a problem in lots of places. So I don’t think you should assume that what happens in America stays in America. We export lots of things for good and for ill — let’s put it that way. So I do worry about that. And I do think that we are at a particular moment now with electrification, where as we ramp up electrification efforts, and as carmakers develop models for their first electric vehicles, it’s an open question of whether we’re going to be faced with a trade-off of basically reducing emissions and increasing traffic deaths, or if we can actually achieve both goals at once. And I say this in, again, another article I wrote a couple of months ago in Slate, all else being equal, an electric vehicle that is identical to the ICE (Internal Combustion Engine) model, is going to be actually more dangerous because it’s heavier because of its battery. It’s also going to be quieter than an ICE vehicle which means that pedestrians are less likely to hear it. And acceleration of an electric drive train is much faster.
So I very much worry that we’re already seeing carmakers advertise these muscle car type acceleration rates for SUVs that are electric, because an electric drive train can let you do that. Why do we want to provide for that? Why is it a good thing to have an SUV with a hyperfast acceleration? That’s not going to make it easier to merge on a highway, it’s just going to lead to at least people confused about how fast their car can go up front, and it might actually lead to, in the long run, just more fatalities or more crashes overall. And as far as I can tell, at least in the US, Congress and NITSA are just — pardon the expression — asleep at the wheel on this. I’ve heard nothing from them. Nothing about managing the shift toward electrification, which is necessary for climate reasons. And a lot of you will have to drive and I get that. But we should be doing this pivot toward electrification of automobiles in a way that reinforces our efforts to reduce traffic deaths, and doesn’t create an unnecessary headwind. There shouldn’t be a trade-off between these goals.
PM: I feel like the expectation that an electric car should look like a sports car is another thing that really comes from Tesla setting the model for what the future of cars, for what the electric car should be. And it’s this car that needs to go really fast, that needs to be like a sports car. And that’s kind of setting the incentives in the wrong direction.
DZ: I don’t know. I can see the argument that Tesla did a really good thing a decade ago by showing that EVs could be cool. We sort of take that for granted now, to an extent in a lot of our circles, where it’s seen, as being a status marker of sorts to have an EV. And But Tesla did break new ground a decade ago, showing how beautiful the car could look and how impressive it could be to roll up in one. Now I think that’s sort of taken as a norm. But at this point, I mean, you look at like the Cybertruck. And you look at all these things Tesla has done in the years since I find it very difficult to take credibly any argument that Tesla as a company cares at all about safety for safety’s sake. They are in a lot of ways, I would argue, strategically sacrificing safety by rolling out features to the public, like full self-driving, that other automakers just wouldn’t do because they think it’s too dangerous. And then by rolling them out faster, and taking risks that others wouldn’t, Tesla can claim to be ahead of the game technologically, even though it really isn’t compared to a lot of its competitors. It can create that image because it’s willing to put things out onto public roads that are riskier. And that’s a business strategy that in some ways is quite effective for Tesla. But it’s one that all of us who have to deal with the after effects, including those of us who are just walking or biking or driving a car that isn’t a Tesla, should frankly, say: What the fuck? This is not a business strategy that should be allowable, and Tesla should pay a price for it, or it should be blocked from some of the more reckless steps that the companies taking.
PM: One thing that I find really fascinating in reading some of your work is that obviously this is a critical technology podcast. It’s critical of the tech industry. But that doesn’t mean that I don’t think there’s no role for technology to play in making things better and making the world better. And you’ve written about how there are actually technologies that we can adopt that can be implemented into the transport system into cars themselves, that can make some of these problems better. But, you know, they’re clearly not the technologies that is getting the tech industry excited that they want to adopt and put into these vehicles. And it’s also something that regulators are kind of averse to, to take advantage of and to mandate to go into cars. What are some of these technologies and why is it difficult to get people on side with them?
DZ: Sure. I appreciate the question because there are some innovations around technology that I’m really excited about. And really briefly one that we’ve been talked about at all is e-bikes and e-cargo bikes. Phenomenal! Love it! And that’s all made possible by the development in battery technology. And turns out when you stick a battery on a cargo bike, it becomes so much more feasible to transport children or groceries — these things are wonderful car replacers. And I am delighted to see this sort of Cambrian explosion of micro-mobility. Micromobility form factors — keep going! But I think with regards to motor vehicles, a couple of technologies that I think could be really interesting. One is actually for buses, automatic traffic cameras that are mounted on buses and can automatically identify the license plates of cars that are blocking dedicated bus lanes.
Because you can have one person who’s holding up 60-120 people in a couple of buses and one of these lanes because they ran out to get a quick cup of coffee and the likelihood of them getting a ticket is virtually nil because police officers don’t really care. And also, they’re only there for two minutes — it’s just two minutes. So this two minutes matter a lot, when you factor it across however many people who are stuck in the buses behind them. So these cameras, which New York City is using, and now Los Angeles, and San Francisco as well, are terrific. There’s studies showing that in New York, that so far, they’re speeding up bus trips. And also this is great. Now let’s try to stick it to drivers. With these tickets, it turns out that the majority of people are just getting one. And once they get that first ticket in the mail, they realize, okay, I can’t do this anymore, and they don’t do it. That’s what you want. You want to change behavior. So I love automatic bus cameras!
Another one that I like a lot for cars, is what’s known as intelligent speed assist. And this is technology that basically is able to identify the prevailing speed limit on a given street or road, and then adjust how fast a car can go. When it is basically being driven on that road, maybe go five miles over the limit or something like that. This is already available. It’s installed on some fleet vehicles. To me, it’s an absolute no-brainer. When you look at some of the catastrophic crashes that have happened, like one in Los Angeles over the summer where someone was going over 100 miles an hour in what I believe is a 45 mile per hour road and ended up in a fireball crash where several people lost their lives. Why should it even be physically possible to go anywhere near that speed on an urban road at 45 miles per hour? Speed Sensor ISA is phenomenal. And carmakers do not want to go down this route. They want to talk about ADAS because it’s useful, being able to at least advertise the potential of you’ll be able to drive really fast in your car. If you notice, by the way in your speedometer how the needle seems to go all the way up to 170 miles per hour often on these cars that you never go past like 80 miles per hour on. Funny that it’s because you feel like you have all this untapped power. Carmakers know that’s great. And they want to maintain it.
ISAs are just a no brainer. And New York City this summer started a pilot I think is very exciting to deploy ISAs (Intelligent Speed Assist) on a bunch of their own fleet vehicles for the city. Data is being collected and analyzed by the Volpe Center, a research arm of the US Department of Transportation. And I’m very excited about the potential of large fleet managers, including public sector fleet managers to adopt isas and sort of spread the good news because frankly, you don’t need every car to have ISA to have an impact. If a given proportion do in a roadway, think about it, every other car is going to have to go the speed that that car with ISA is setting. So you just created an obstacle toward reckless driving with only say, I don’t know, 20-25% of vehicles having it. So I love that those are a couple examples of technologies that do get me excited, even when I think AVs and ADAS, in different ways, the hype very much exceeds the reality.
PM: No, I love that I think it’s fascinating to look back like in Peter Norton’s book as he describes the fight for speed limiters all the way back in the 20s. And to see how we’re trying to still make that happen today. I think e-bikes are fantastic. I know there’s been some kind of back and forth in some European countries around them. But I feel like especially in North America, they make a ton of sense where we have these distances that are a bit larger, because a lot of our environments are built for cars. And so they’re kind of perfectly suited for the urban environment and making it easier for people to get around without a car. And then of course, I love hostile bus technology — bring it on! That sounds fantastic. Giving tickets to cars who are blocking the lanes, it’s great. A final question as we kind of wrap up this fantastic conversation. We talked a bit about the incentives of politicians and regulators in approaching these questions, whether it’s road safety, whether it’s technology in transportation, and how they really haven’t gone far enough, they haven’t kept up with these issues to ensure that they are ensuring that the roads are safe, that actions are being taken to reduce deaths, to make the transportation system better for people. Do you feel like there is starting to be a shift in that direction? Personally, I’m not a huge fan of Pete Buttigieg. But he seems to be talking about some good things. I feel like there’s been discussion in some of the regulators in the United States around taking more action on autopilot and Tesla and bringing in more regulations on automated driving systems, etc. Do you feel like we’re starting to go there? Or do you feel like there’s still a lot more pressure that needs to be applied for them to take the actions that are really necessary?
DZ: I think that we are making progress, but still, there’s a lot more pressure that needs to be applied. So little bit of both. I actually think Secretary Buttigieg really does care about this stuff and would like to make a lot of progress on it. Maybe I’m a little bit more bullish on him than you are in that way. But he’s really hamstrung in a couple of different ways. He’s hamstrung politically by a Congress that is not interested in picking fights over SUV size or on limiting the speed that a car can go. I actually interviewed Buttigieg when the National Roadway Safety Strategy came out in an interview that was in Slate, and I asked him about whether vehicle size reduction for SUVs and trucks was a national priority, and he danced so well, I couldn’t even tell that he was not in any way shape or form answering my question. He says a lot of the right things. He was just recently saying autonomous vehicles feel like they’re always 10 years out. He’s talked about how we have a true national crisis in roadway safety, which I think is constructive. But he’s facing a NITSA that is, in some ways, I hate to say it, but I do wonder if they’re, in part, at least captured by the auto industry. And you see how many people like the former NHTSA administrator who’s now doing gov relations for GM. And this revolving door phenomenon happens so often now it’s really problematic.
So even having said that it’s positive seeing that there’s rumblings of taking action against Tesla in a more serious way, when frankly, NHTSA was just again, asleep at the wheel for years talking about the promise of autonomous vehicles and wanting to keep regulations from getting in the way of their development, rather than creating safeguards to ensure Americans are safe as these new and untested technologies end up being used on public roads. So I will say I think this only gives me hope. It’s that NHTSA is, whether it likes it or not, is facing a more engaged group of Americans who are just pissed off about the status quo. And the rising number of traffic deaths and the recognition that part of the cause of the road traffic crisis in the US is NHTSA’s inaction around vehicle size, around infotainment. There have been a couple of instances in the last couple of years, where there’s something that ends up in what’s called the Federal Register. It’s sort of a proposed rule that NHTSA is putting out there, first, for revisions to something called the MUTCD, which is sort of the Uniform Code for signage that has for a long time stood in the way of cities taking steps like reducing speed limits, or painting crosswalks in bright colors. These are things that in MUTCDs is somewhat frowned upon, and creates big headwinds. There were revisions that MUTCD proposed and then later revisions to the crash safety ratings program known as NCAP.
And both in those instances, over 10,000 responses were submitted by the public each time, predominantly from people just saying: You, NHTSA, need to be far more proactive in supporting street safety and in regulating oversized cars. And I have heard from sources at NHTSA, that they’ve just never seen this level of public engagement. And they aren’t quite sure what to do about it, or how to handle it. They’re just sort of processing. And I find that really constructive. Because I think five years ago, I remember thinking how we really have a problem that we have these local street advocacy groups or bike lane advocacy groups that don’t really engage at the national level even aren’t even aware of what’s going on there. And I don’t think that’s true anymore. I would never write that article, which I thought about writing a little while ago, because I do see much more engagement now in the US. And awareness of ways in which federal regulations and federal policy are leaving Americans in danger when they travel, especially if they’re not inside of a car. And that is a trend, which is still nascent. But it does give me hope.
PM: I think that’s a good place to leave it. Because if the tech companies and if the prospect of autonomous vehicles and these other supposed solutions, maybe left people waiting for a few years to say: Okay, maybe this is going to solve it, it’s become pretty conclusive that that’s not going to solve the problem. So there’s more reason than ever, for people to get engaged and to finally start demanding the political fixes that are available right now that we can take to actually solve these problems. David, it’s been great to speak with you. I really appreciate you taking the time. Fantastic conversation. Thank you so much.
DZ: No, it’s really been my pleasure. I’ve enjoyed it as well.