Could Amazon’s Zoox Beat Tesla and Waymo in the Robotaxi Race?
32:50 Share

KPMG makes the difference by creating value, like developing strategic insights that help drive M&A success and embedding AI solutions into your business to sustain competitive advantage, or deploying tech-enabled audits to deliver more accurate and transparent outcomes. KPMG drives brighter insights, bolder solutions, better outcomes.

It's how our people make the difference every day. KPMG, make the difference. Learn more at www.kpmg.us/insights. Hey Mims, would you ever get into a car without a steering wheel or brake pedals? What is this, some kind of crazy YouTube stunt?

That's life in San Francisco. It's actually what Amazon's robo-taxi company Zoox is doing, building a bunch of vehicles that look more like living rooms on wheels. Well, I can't wait to try out the couch, put my feet up, and we're going to hear more about Zoox's plans from its CEO and why she didn't ax this radical car design when she took charge of the company. That's next. ♪

Recently, my co-host Tim Higgins took a ride around San Francisco in a vehicle made by Zoox, Amazon's robo-taxi startup. Yeah, we'll ride forward. Why not? Okay. It's your first time, so... Yeah, oh, should I sit over there? No. Unlike some of its rivals, Zoox's vehicles don't look like traditional cars. They're sort of like toasters on wheels...

or maybe a theme park ride that hopped off the rails. They don't have brake pedals or a steering wheel, just two rows of seats facing each other. It's almost like a lounge. It was actually the London cab was the inspiration. You can go ahead and press start for your ride. Yeah, there you go. And yeah, the whole thing is we don't want you to think about driving. We want you to think about being transported. Tim hitched a ride with Zoox's CEO, Aisha Evans. Before this, she was a senior executive at Intel.

And after she was brought on board at Zoox in 2019, she sold the startup to Amazon for $1.2 billion.

She's a skilled manager who in many ways has met this moment in autonomous driving, a time when what she calls patient and smart capital is required to get to the end of a self-driving race that has proved to be a marathon. It runs into, well, what happens if it approaches a driving situation that it doesn't understand? It just did, by the way. Well, right, this one it understood. It saw some people opening their doors, so it went around it. Yeah.

In an industry where scale could prove to be everything and safety is paramount, Evan's style of calm, steady-as-she-goes leadership could hardly be a bigger contrast with that of a competitor, Tesla CEO Elon Musk.

But Amazon has put a lot of money into Zoox, which will be starting a commercial robo-taxi service in Las Vegas this year. This is not an industry where companies can afford to be cavalier with safety, and Evans knows it. If you do a prompt on one of the best AI engines today, you know, it's 99% right. Maybe it's annoying, but that's okay. In our space...

If it's wrong 1% of the time, that is not okay. You just can't. That's not acceptable. The bar has to be way higher than that. From the Wall Street Journal, I'm Christopher Mims. And I'm Tim Higgins. This is Bold Names, where you'll hear from the leaders of the bold name companies featured in the pages of the Wall Street Journal. Today we ask, what was it like taking over a self-driving car startup at a major turning point in the industry? Aisha Evans, welcome.

We want to get into Zooks' plan for deploying robot taxis, but first, I think it's worth taking a little journey back in time to about how you and Zooks got to this point. It's almost exactly your sixth anniversary taking over Zooks as CEO. You're

come from Intel before. And your arrival was really a turning point for the company. But it was also, as we would see in the months that followed, a turning point for the driverless car industry that had been up until 2020 in overdrive. Very frothy, as we like to say here in Silicon Valley. But then the cash was disappearing. So I'm curious, when you took over, what was your first priority?

My first priority, like I would call it priority zero, is you're coming in from a very established company. I told the team that I wanted them. I knew it would be OK when they would not remember that I wasn't here at the beginning. And so talking to a lot of people, understanding what's really going on, getting a feel for the industry.

After that, I went to priority one, which was it takes a lot of capital and a lot of patience to do this. And so we needed to solve that. What was your strategy for pivoting Zooks' strategy with these employees? Many of them had begun with this scrappy startup with a very charismatic founder who had really bold vision for where he was going to take Zooks. So what was that like for you?

It was lonely, but it was, look, I have a slightly different view. When you think about, essentially, these are computers on wheel from a technology standpoint. They have to be safe, not just 99% of the time, but 99.5%.

Pick your number of nines. We think it's four plus. That's a lot of work. You're bringing also a variety of experiences, background. One day you're talking to kind of an AI genius at the forefront of things. The next day you're talking to, you know, a car mechanics kind of person and then a battery person. So bringing everybody together and saying congratulations, because you have to have some audacity.

And boldness to go after something like this. And so thanking them, congratulating them, and then saying, okay, but what got us here won't get us there. And so this is going to take time. We need to slow down, stop with all the talking to the press and making bold claims and don't

go back into the building and sort of figure out, A, the capital and partnership situation, and B, build a roadmap that is incremental, that is deliberate, and use a lot of analogies and know that it's going to take a while. This is almost like the beginning of aviation in a way. And one of the big things you did almost immediately was...

look at that capital table, I imagine, and say, oh, we're in trouble, right? I mean, and that puts you on a road, a path to eventually being acquired by Amazon. That is absolutely correct. And look, it's not just you need money. You need what I call patient and smart capital, meaning people who understand deep tech and

people who can see a bold vision and the steps towards there, people who are close to the compute industry and who are not going to oscillate too much and who can afford to take a bet like this. And when you do that math and calculus, there aren't that many companies. We're thrilled that we went with Amazon. And even though the press wasn't kind to me at the time, I am very, very grateful and proud that we made that move. It freed us up to go after the work.

as opposed to making claims and demos. Because demos are very easy in this space. They are getting even easier. But getting this thing to work, to be safe, to operate, to wow the customers and build a business, very different movie. So speaking of wowing the customers, one of the things that's so distinctive about Zoox when you first encounter it is the fact that your vehicle's

look like they are from maybe the near future. It's not a Cybertruck, but it's a lounge inside with two benches facing each other, right? It could go forward or backward. It feels like a happy, benevolent future where everything runs on wind power and there are no more car accidents. Obviously, that's deliberate. But what's the thinking behind that?

First of all, if you step into that vehicle when we're fully commercially deployed and you think about driving, we failed. We want you to feel like you're being transported. Then there is at scale, the level of safety, the level of control, the level of redundancy that you need. Negotiating with somebody who might be a company that's a traditional OEM that's in a different mindset will take more capital.

We'll take longer, more meetings, more back and forth and more detours and really take away from solving what's really a very hard problem. OEM being lingo for carmaker out there in the layman's world, which is I think you're kind of getting at something we're going to get into, the idea that you're actually making these cars yourself, unlike Waymo, Google's effort, which is.

kind of taking a car and putting its equipment onto it and converting it over. This is ground up for Zoox, which was a pretty radical idea given how much capital was going to be required, right? You were not only developing the self-driving brain, but you were also developing a car. And that is very audacious and very grand. So I guess I'm curious, what...

is that timeline or how close to having paying customers in the vehicles are you? We're getting closer and closer, but you know, paying cost the journey. It's funny. We just had a meeting yesterday and I was talking to some of the leaders and I'm like, look, it took us 11 years.

To put this robot taxi that we love, even to this day, when I sometimes I'm driving around and I see a couple of ways by, I giggle. I'm happy. And so but that took 11 years. Let's not have a disrespect for the journey to commercialization and customers. That's going to take some time. So we will dip our toes.

into it later this year. We'll learn a lot, but it'll be small scale because again, that's a new journey and the next couple of years are really exciting for us. And so right now, what is, you're currently at the what? What is it currently available at? Because I'm in San Francisco and I've seen these vehicles on the roadway. What's going on?

We're testing. We just started testing in San Francisco in November and also in Las Vegas from our depot, which is southwest of the Strip, all the way up the Strip. We finally have friends and family riding.

which is really cool. And of course, select VIPs, whether it's press, regulatory and execs and what have you. And so we'll go now into the phase where they'll be able to ride without a Zoox host and push every single button

And then we'll shortly add what we call Zoox Explorers, which will be people that we will sign up and will have the ability to ride whenever they want. That period where somebody will be in the vehicle and I don't even know who the hell they are. And they'll give us feedback. So all that is planned. So we're at the end of tiny and entering the beginning of small and then paying customers later this year.

We've heard how Zoox is aiming to be a transportation company that makes you forget that cars even existed. But making that happen will require solving a lot of hard and occasionally gross real-world problems. What happens if somebody pukes in the car? Oh, this is, especially in Vegas, I suppose we could be prepared for that. What happens in Vegas stays in Vegas. Maybe it stays inside the back of a Zoox.

How will Zoox handle the mundane realities of operating a fleet of vehicles that people are prepared to occupy, if not drive, like they stole them? Stay with us. ADP imagines a world of work where smart machines become too smart. Copier, I need 15 copies of this. Printing. By the way, irregardless, not a word, Janet. Yeah, I know. Page six should be regardless of or irrespective of. Just print them, please.

If it were a word, Janet, it would mean without irregard, which is... Copier! Switch to silent mode. Let's put a pin in it. Anything can change the world of work. From HR to payroll, ADP helps businesses take on the next anything. So as you're moving toward paying customers, I'm really curious even further out how this business works.

Just on a fundamental profit and loss level, what is your kind of unit basis for this? Because obviously with Uber, the reason they have worked as a business is they don't have to own cars. They don't have to have the risk of owning those vehicles. You know, all of that's taken on by their drivers, but you have to build vehicles, own them, maintain them.

You know, even if you're not paying wages to a driver, it seems like that might be hard to pencil out. So how ultimately do the unit economics on this work? So I'm going to be really careful. First of all, I'm going to back up the boat a little bit. You've both said twice build. So let's talk about what we do with the vehicle. We architect and design it.

And then we work with a global supply chain. And then what I would call 20 big Lego pieces arrive at our factories. We have a prototype factory that they arrive at. And then from there, we do assembly, test, integration, bring up, and then ship off to depots where they will operate from. So I want to be very clear. There's no gigafactory here. There's no any of that stuff. So in our case, volumes are low.

Because it's our fleet, we know it super well, the robot, because we designed it, we architected it, we put a lot of features in it thinking already about operations. That's, again, a first principle thing. The robot pays for itself very quickly.

at the current prices that folks are used to with the ability to bring down the pricing over time as you're going to different phases. I was at Intel beforehand. I don't know a lot of hardware businesses that have a lot of software and algorithms that pay for themselves overnight.

And if anybody pretends that they know how to do that, I will just tip my hat to them. There's going to be a buildup where you'll need profitability all the way to margin and all the way to the phases of free cash flow. That will take time, but the opportunity is immense. And we also think that the time will be expanded because there are a lot of folks who don't ride today that will ride in the future. I was telling Tim I'm a soccer mom.

And now I'm lost with myself because my daughter is in college. And I'm like, why do I have so much time? And I know I have so much time because I am not driving up and down the peninsula in every nook and cranny, dropping up and picking up kids for soccer. As a minivan driver and owner, I definitely identify with that. And it's also been amazing to see the pickup of Uber's new product for teens, right? Which, of course, bolsters

your thesis now, which is there's a huge potentially untapped market. I mean, it doesn't take much imagination. But I think that one of the things, and we've kind of been alluding to it, there's other costs. It's not just the technology. There are other challenges that you have going on there. I think conventional wisdom might be that

You just flip a switch and the robot goes to work. But it's not that simple, I'm betting. No, it's not. It's more like aviation than it is like the robot vacuum going around your house. Look, I mean, we have to have a lot of respect for this. This is a very big, heavy robot that is roaming around and driving amongst other human drivers, by the way.

And when you look at the math, whether it's collisions, whether it's injuries, whether it's fatalities, when you look at the ability to not make mistakes, obviously, or at least minimize your number of mistakes, but also minimize

react to other people's mistakes, it's an infinite long tail. And so we've been very open and very transparent that we have a teleguidance center, almost like an air traffic control tower type thing, where there are operators that the robots can call for help.

If it sees a situation that it knows it doesn't know how to handle, imagine like a super, super, super, super unstructured construction zone or some blockage that it's not sure how to maneuver around. It has the ability to call and say, hey, over this period of time and over this distance, I see something that doesn't make sense to me. I either need a permission to break a rule or I

it can receive an alternate trajectory via breadcrumbs that it is still in charge of deciding it's safe enough to maneuver. And so obviously the, as I, my team's going to kill me for saying this, but I've always said, look, if I'm going to need one or two people for each robot, we might as well

not do this because then it's already being served today. Right. It doesn't work if you've got one person who's watching one car and then another person who's cleaning it. But as you scale up as a business where one person is watching X amount of cars and one person is taking care of X amount of vehicles, you start to get some scale and some savings. And that's where you see the business going forward. Absolutely. Absolutely.

What happens if somebody pukes in the car? Oh, this is especially in Vegas. I suppose we could be prepared for that. What happens in Vegas stays in Vegas. Maybe it stays inside the back of a Zoox. There you go. You can use that tagline if you want for your advertising. I will give you credit for it. So basically what we do is we're monitoring the inside of the vehicle and that vehicle would be pulled out of base and either out of the fleet or

and either go to a hub or go back to Depot. Given what your goals are, it's hard to see a connection to Amazon other than as patient capital. But is there a connection there? I mean, they're fast becoming the world's biggest logistics company. I could see some potential overlap. But are there any ideas about that? Have there been in the past?

They are more than patient capital. They've also done a lot of different businesses. There's a lot of experience there. And so they also give us really good advice. They challenge us. The first order of business is to deliver what we've said, because, again, it makes sense that if it's this hard to solve the problem and we all drive on the same roads, right? The trucks, everybody drives on the same road. So our first order is to deliver this business and build it, grow it.

We've had conversations about other things we could do. Obviously, delivery is one, but that's a tough problem to solve. And it still goes through solving the first problem to begin with. Who do you see as your biggest competitors as Zoox at this point? Is it Waymo? Is it Tesla? Is it just Uber with drivers? So I'm a little different than most people. This is a big industry. I try and have some respect and humility here.

And so I have called them fellow travelers because I was not confused that I was going to be alone or we were going to be alone doing this. That to me would be delusional, at least the way I think. So it's fellow travelers were also indirectly in the safety business. What either of us does affects the other, whether we like it or not. So the way I would answer your question is that we've always thought that

There would be a wave of consolidation past the demos, and it would be three to five players in America, probably up to 10, 15 in China. And then Europe will see what happens, as is the case in many different industries. Three in the U.S., and you think maybe Waymo, Tesla, and yourself. Who's Coke? Who's Pepsi? What are you? Oh, I'm...

I'm trying to. I'm an excellent chai latte. Well, I guess I'm curious on that field of competition. I get what you're saying, and it's still probably a few years away before the competition gets very intense, and we all know it will. But I wonder...

How do you compete against the likes of a Waymo or a Tesla? Because in a lot of ways, if the robot is driving, it's almost a commodity experience. Is it a race to the bottom or is it a race to create the most luxurious box on wheels? So we look at it as the customer experience and the full end-to-end experience, including customer support, by the way, is going to be key. We think that our experience is magical. I mean, this is your space.

You don't think about driving. You're not like, oh, the wheel is moving. Is that okay? Or anything like that. It's like whatever you were doing before you entered, you can continue doing it. I've had some people crank up the music and start just rocking out there. I've had some people just meditate. So the personalization, the simplicity, all of those things are what we think we will offer. And we have a parent that knows a lot about operations.

that will teach us a lot about operations too, which is key because little things like that matter. Pick up and drop off, wait times, when something goes wrong, how do you react, how quickly do you react? And so we think that will be up there in terms of that. It's almost like, yeah, it's a first class seat, but you paid coach. You paid for coach. We just heard about how Zoox will tackle the nitty gritty of rolling out a fleet of robo-taxis.

The company faces the additional challenge of rolling out a vehicle that looks like almost nothing else on the road. This almost wasn't the case. And the history of the company's funky ride is a tale worth hearing. All of my mentors, friends, sponsors would be like, the first thing you need to do is get rid of that vehicle. That is the first thing you need to do. Those people are crazy. That's next.

Your data is like gold to hackers. They're selling your passwords, bank details, and private messages. McAfee helps stop them. SecureVPN keeps your online activity private. AI-powered text scam detector spots phishing attempts instantly. And with award-winning antivirus, you get top-tier hacker protection. Plus, you'll get up to $2 million in identity theft coverage, all for just $39.99 for your first year. Visit McAfee.com. Cancel anytime. Terms apply.

You mentioned that you think in the U.S. there'll probably be three big players here ultimately after the consolidation is done. Obviously, Waymo is going from strength to strength. Tesla always feels like a wild card to me because, of course, they have an incredibly deep bench of talent, but the sort of revealed reality of their...

you know, so-called full self-driving software, which is on the road in personal vehicles, is always that it can be a little erratic in ways that certainly folks find unnerving. But it sounds like you think that they're credible. Why is that? Like, can you kind of square that circle for me?

It's just I mean, it's not my I don't like it when people criticize Zoox and tell us what we're doing or not doing, even though they have no internal information of what's on what's really going on in here. So I'm not going to treat somebody or another company a way I wouldn't want to be treated. I don't know what they're up to.

I know what they've said. And obviously, like the same is true for us. The proof is in the pudding. I mean, when I arrived in whenever it was six years ago, the number of people, I mean, all of my mentors, friends, sponsors would be like, the first thing you need to do is get rid of that vehicle. That is the first thing you need to do. Those people are crazy, blah, blah, blah, blah, blah, blah. And I was like, wait, what about I go in there?

and seek to understand and get a little bit more information because, you know, I'm not one that tends to think people are stupid. So I have to take them at their word. And it's the same bar we have. We can say whatever we want, but if we don't show up with a driving thing, then we have to shut up. At this point, Tesla is saying,

A robot taxi they plan on deploying this summer in Texas. Another competitor, Uber CEO, told our colleague Joanna Stern in Davos that he didn't see humans being replaced in scale behind the wheel for at least another 15 to 20 years. Is that the time when robots replace humans behind the wheel or is it sooner?

I want to be careful because that can be fairly inflammatory. Remember, we are creating a lot of jobs at all skills levels and that's not going to stop. So I think we're just having humans do different things than being behind the wheel.

Look, it depends on how you define scale. If we're talking about covering the entire United States and even venturing internationally, yeah, I think that's a reasonable timeframe. But along the way, city by city, state by state, you will see progress, progress. Again, if you...

have the chance to be in San Francisco. It's pretty amazing. If Ben and Bertson and Takedra and Dimitri, please say thank you after this podcast. Literally, you whip out your phone, you download Waymo, and

few minutes, you're good to go. And you say where to and you enter the information and off it goes. I like that you believe that the co-CEOs of Waymo are avid listeners of this program. I believe they probably should be because you are also getting at an issue that I think that they are dealing with over at Waymo, that Tesla will have to deal with, that traditionally in this country,

we have regulated the car at the federal level and we have regulated the driver at the state level. So the big question becomes when the car is the driver, where does that regulation fall? Right now, it appears that it's going to be at the state level, though there is a push for it to be federal. What do you think needs to occur there? I mean, the first thing is a regulation.

I can wish for many things, but I have to first learn to live with what is currently happening. And so we work super hard to be transparent, have no surprises and really in a collaborative manner at the federal level, because we have the vehicle, at the state level, because they basically regulate the driving and the driver, and also at the city level, because it's their streets.

And different cities want different things at different times. So that's how it is right now. And it's at least possible. None of us are really going to 20 cities overnight. So I think for now it's okay. But as we start to enter the medium to large market,

scale as an industry. I obviously hope that we can deal with less than three regulatory agencies. And so it'd be nice if there was some unification. But, you know, beggars are not choosers. In the meantime, we have what we have. And we're lucky that there's even a way to do it. It's been clear throughout the evolution of this industry that the gating factor is safety.

And obviously Uber crews had incidents that, you know, led them to shut down or sell off their programs. What, if anything, have you learned from the failures, the safety failures of those programs? Learned a lot. I joke that sometimes there are some advantages to being number two. Look,

One of the things with Zoox is because we because our robot taxi is what we believe is the final sort of hardware solution at scale. There is a luxury we don't have. I can't say, oh, don't worry about it. If something happens, we'll just send a driver to drive it away. I also don't want to run a business where we're, you know, flooding the city with tow trucks to go rescue these robots. And so safety matters.

It's totally foundational to everything we do, but I'll be honest, no matter how many scenarios you think about, you always find out another scenario. And so we've learned to really think about not just safety in don't hit something, don't injure, don't have a fatality, but also in what happens when you're standing still and there are other, what reactions are you causing that

Because this thing is different from what human drivers are. And then every time that we see something happens, we basically add to our bank and to our regression and to our structured testing. And then last but not least, in all of our computational for safety models, we assume that there's something we don't know. And when that happens, how do we want to react?

And then the third vector is, and sometimes it's hard, super transparent with everybody. The first responders, the PDs, fire, the cities, at the federal level, at the state level, super transparent so people know what to expect. And so we build a little bit of a bank of credibility. We never want to leave them with the impression that we didn't tell them what they were asking about fully.

Well, buckle up. The future is here. Aisha Evans, thank you. And see you out on the road. My pleasure. And I hope to see you soon for Zoox Rides. And that's bold names for this week. Michael LaValle and Jessica Fenton are our sound designers. Jessica also wrote our theme music. Our producer is Danny Lewis. We got help this week from Catherine Millsop, Scott Soloway, and Philana Patterson.

For even more, check out our columns on WSJ.com. I'm Christopher Mims. And I'm Tim Higgins. Thanks for listening.