The Self-Driving Experiment 35:00

When you're running a small business, sometimes there are no words. But if you're looking to protect your small business, then there are definitely words that can help. Like a good neighbor, State Farm is there. And just like that, a State Farm agent can be there to help you choose the coverage that fits your needs. Whether your small business is growing or brand new, your State Farm agent is there to help. On the phone or in person. Like a good neighbor, State Farm is there.

Silicon Valley Bank is still the SVB you know and trust. The SVB that delivers human-focused, specialized lending and financial solutions to their clients. The SVB that can help take you from startup to scale-up. The SVB that can help your runways lead to liftoff. The only difference? Silicon Valley Bank is now backed by the strength and stability of First Citizens Bank. Yes, SVB. Learn more at svb.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.com.

There's this video that Tesla put out way back in 2016 to drum up excitement about a new feature called full self-driving. The video opens with a black screen and a disclaimer that reads, The person in the driver's seat is only there for legal reasons. He is not doing anything. The car is driving itself. A guy gets inside a car and it pulls out of a driveway as the Rolling Stones begins to play in the background. The steering wheel spins on its own as his hands hover around it.

The video shows the car navigating flawlessly through Palo Alto for about three and a half minutes. The song's energy makes a leisurely drive through the suburbs feel important. At the end, the car drives up to a Tesla office building, drops the guy off, and parks itself. Even now, this clip is exciting to watch. I had never seen anything like it at the time, and it gives every impression that this Tesla can fully drive itself. But here's the thing about that video: it was all staged.

Behind the scenes, the entire route was pre-programmed. And still, drivers had to take control of the car several times. It even crashed into a fence while it was trying to park. But just a few months before this, Elon Musk had gone on stage and claimed to have cracked the code. I really would consider autonomous driving to be basically a solved problem.

Like a Model S and Model X at this point can drive autonomously with greater safety than a person right now. Musk said this seven years ago, and it wasn't close to being true. In the years since, he's continued to make this claim. All right, and we have full self-driving in beta. Anyone here use the full self-driving beta?

The promise of autonomous cars is convenience, like having a chauffeur that never gets tired or drunk or distracted. You can really get a feel for how much full self-driving is going to transform society. The car will be able to take you anywhere you want. It's also the promise of safety. More than 40,000 people in the U.S. died last year in car crashes. With ultimately 10 times safer than if you were driving it yourself. And it's going to just completely revolutionize the world. This is one of those things that comes along very rarely.

So what about today? Can Tesla really make a car that drives better than we can?

This is Land of the Giants. I'm Patrick George. Today we're looking at one of the biggest and boldest moves Tesla has ever made. Its foray into self-driving cars. Over the past few years, no car company has pushed this technology more aggressively. But it may have also pushed things too far and too fast. And self-driving is where some of Tesla's wildest promises have yet to materialize, and may not for a long time. Welcome, guys, to the Tesla.

- You ever run one before? - Oh yeah. - I've driven many of them, but never one with FSD. That's my first. - The Tesla tour guide you're hearing is Omar Shakir, our colleague at The Verge. On a recent sunny afternoon in Jersey City, my producer Zach and I piled into Omar's Model 3 equipped with Tesla's full self-driving software, FSD for short. - And I'm just gonna select Whole Foods, which is like north of the Hoboken. So I'm just gonna go ahead and just start driving, and then I'm gonna go ahead and press this stick down twice, and now we're on full self-driving mode.

We wanted to see how FSD handled the complexities of city driving, to give FSD a driver's test. So think of this Tesla as an eager teen trying to get its license, and us as the scary DMV instructors here to grade its driving abilities. I'm not doing anything other than just keeping my hands where it should be. Alright, so we have some people at the crosswalk. I have my foot hovering over the brake. I have my hands hovering with the steering wheel so that I...

can make any quick decision I need to. Omar has to be in the driver's seat, ready to grab the wheel at any time because Tesla requires it. So FSD is technically a partially automated system. FSD uses a combination of cameras, sensors, and AI to get around. It's meant to be able to handle city driving like this on its own.

Despite it being the middle of the workday, there were still tons of things the car had to navigate. Potholes, construction, red lights, pedestrians merging, aggressive jersey drivers. A red light at a four-way intersection was the Tesla's first real test. As you can see on the screen, it's observing people, you know, crossing the street. It's observing cars coming on both sides. Now the car is creeping forward and it's checking again. And the car looks like it felt uncomfortable to make the move.

And now it sees a person coming and looks like it paused again. It's definitely erring on the side of caution. And now watch the lights just changed and now it stopped attempting. We just missed that light. But look, look at this. It's trying to creep forward even though it's a red light. I just stopped the vehicle myself. It was trying to creep forward on a red light, which is not a good thing. Omar makes the turn himself. Right after that, he switches FSD back on.

Just in time for the Tesla to make a mistake changing lanes. Look at this, it tried to pass on the left, but now it wants to get back into the right lane.

So let's see if any of these cars let us go. And so I realize it messed up, it's in the wrong lane. And right now it should just change lanes. I think it's a little bit in between decisions right now. I do wonder what's going to happen once the light turns green. And my instinct is telling me that I might need to take over in a little bit. But we'll see what happens once the light turns. Sure enough, Omar needs to take over again to get us into the correct lane to make the turn.

All right, I have taken over because I don't want to get honked at. Yeah, I think you did the right thing. We're going to go left instead of right where you wanted us to go. So far, let me look at my clipboard. It's not going great.

It took me a speed bump very quickly. It took that speed bump really quickly. I'm flashing back to about 20 years ago, and I feel like I'm my dad sitting in the passenger seat being like, now watch out for that car. What are you doing? Why are you going so fast? What gear are you in? Oh, man. It does not swerve for potholes. There's a lot of potholes, and it does not try to get out of the way of those. Funny thing is Elon Musk had once promised that the system will detect potholes and avoid them. From my experience, it does not do that.

It almost takes more mental energy to babysit the car than it does to actually drive it. Bumpy! Man, it took that speed bump. It wasn't all bad, though. We did see flashes of FSD being a safe and competent driver. Stopped for another pedestrian, stopped at a stop sign, and it continues. It did what it was supposed to. Yeah.

We should also mention that Tesla has a related driver assistance system that helps handle highway driving called Autopilot. Think of it as a souped up cruise control. It comes standard on every Tesla, and Tesla drivers repeatedly told us it works pretty well. In theory, Autopilot is a close cousin to FSD, which is designed to handle everything outside the highways, like city streets, where driving is a lot more complicated.

Is he yelling at me? The cop's yelling at you. A traffic cop was yelling at us to move up, something our Tesla couldn't hear or understand. We just got yelled at by police officers. So I just hit the accelerator a little bit. I haven't shut the system off, but I did hit the accelerator for it to move forward. While we were stuck in traffic, Omar told us a little more about experiencing the beta part of the FSD beta software.

collects data as we're driving and it feeds back to like the mothership and that makes autopilot better based on this data, correct? That is correct. I am basically driving for Tesla without getting paid.

How has yours performed to date since you had it on? It's a mixed bag. It's definitely been improving. And that's what's the funny part of accepting this beta software is that you as a user of beta software sort of have to accept that this is not going to work as advertised until a good while. In a 90-minute drive, Omar had to grab the wheel and take over every few minutes. This is like 20 full-cell driving on hard mode in terms of where we are right now.

But that's exactly what is supposed to be capable of this. We can't all live with our Teslas in places that have absolutely no civilization. And that's not realistically where people own Teslas. There can be a false sense of security when you're using this because sometimes it's actually really impressive. But you need to think of it more like a hat trick. For me, I feel like I'm more of a conductor. When I'm using full self-driving, I'm conducting a vehicle.

So I'm doing the checks and balances of everything. And I continue to be impressed with some of the decisions that you can make. But at the same time, you know, this is still beta software and it's not quite there yet. Despite some impressive moves, Tesla's FSD did not pass our driving test. It just doesn't feel fully ready for the road. And it definitely doesn't make good on Musk's promise of being, quote, safer than a human driver.

But critiquing Tesla's FSD is tricky because it's still technically in beta. Plus, at any moment, Tesla can say a bug's been fixed by pushing out a new over-the-air software update. In fact, there's already been an update since our drive, but that doesn't mean the updates are always improvements. Some of those

updates or regressions actually. And they were just being kind of reckless. This is Jon Bernal. He worked at Tesla from 2020 to 2022 running tests on FSD and advanced driver assist technology. Every time we would push a new update every two weeks, they would tell us how incredible the progress has been and how that update specifically may have helped lives, etc. Early on, Bernal was so excited about his work that he went out and bought a Tesla for himself and began testing FSD beta in his free time.

We've noticed that the Beta is very cautious, especially around pedestrians. So now we're here to put us in the wrong lane. He started posting videos of himself on YouTube testing FSD. Many of them were positive, showing off the car's impressive handling maneuvers. But he also showed the times when it wasn't so smooth.

Now, whoa. So it thought we could go straight through there. There was a fence and a curb. A wall. Yeah. This was made about two years ago. We went to Oakland and we tested the system and it was very much an infancy. And because of that, it did a very bad result and almost caused the vehicle to T-bone other vehicles three times and almost went through a fence. Wow. Okay.

Because of that video, Tesla reached out to me internally and like, hey, you got to stop putting bad press out there. You have to edit it to show it in a positive light. So I did that for a little bit. We reached out to Tesla to give it a chance to respond to this and everything else we talk about in this episode. We never heard back from anyone at the company.

Bernal kept posting test drives on his YouTube channel. But after a few months, he posted a video that showed his car making a pretty big mistake. Changing lanes. Oh, what the fuck? Oh, shit. Shit, we hit that. We actually hit it. We hit it. Wow. I'm in San Jose on FSD Beta 10. We unfortunately hit a green boulder and crashed the vehicle at 12 miles per hour while engaging FSD Beta.

Not a deadly crash. No one was hurt. But the car drove into something in plain sight.

Bernal posted the video showing the collision. A few weeks later, major news channels picked it up. And then Bernal got fired. I thought I was building this incredible product. I went and bought Tesla. I got the software. I tested it. It was a fun gimmick for a week. But then I very quickly realized that it was absolutely nowhere close to ready or what the public thought it was based on Elon's tweets. And this was probably about 2020, 2021 era.

Calling it beta lets Tesla off the hook a bit. It gives the impression they're still testing everything. But how beta is it really if anyone who wants to use it can just simply buy it? These aren't certified testers. They're just customers. And as of right now, Tesla says there are over 400,000 cars with FSD installed in the U.S. and Canada alone. Plus, it's rolling out now in Europe. You know, I think that this approach of just giving it to anybody that has the money to pay for it is reckless and irresponsible.

Sam Aboul-Samet is an expert on automated driving and emerging technologies with a background in engineering. He's also spent years developing safety systems for vehicles. Just because you have the ability to pay $15,000 for full self-driving capability doesn't mean that you are qualified to actually evaluate it and test it, especially on public roads. Keep in mind, this is a huge experiment that is being done in public.

You might be wondering, how is this even legal? Well, for something to be illegal, there have to be laws and regulations in the first place, and there really aren't any here at the federal level. There's no driving test for FSD, there's no pre-approval system, there's only investigations and recalls after the fact.

Instead, the laws get punted to the states, and only around half of US states have laws around autonomous and partially autonomous cars. So regulations on this are murky and inconsistent at best. New technology like this, because it's so new, rarely has proper oversight. And we see this sort of thing a lot when it comes to tech companies. The move fast and brake thing, shipped is better than perfect ethos. But that's usually with a buggy app on your phone, not the safety of a driving experience.

Because the thing about a 4,000-pound high-performance car is that it can literally move fast and break things. "Well, we now have video of the moment that a Tesla in self-driving mode started breaking unexpectedly on the Bay Bridge, causing eight cars to pile up." "We're learning more about the driver killed while riding in his self-driving car." "The first ever driverless car death." It's not the only investigation underway involving Teslas and the car's self-driving features.

At this point, there have been 736 crashes and 17 deaths linked to Tesla's driving assistance systems. That's according to The Washington Post. For Tesla, the assurance is that eventually FSD will be incredible, that this is all just one over-the-air update away from working seamlessly. Samabu Al-Samad believes this is false advertising. It's a lie.

because what they're selling is probably never actually going to be capable of being fully automated or full self-driving. Here's why. Tesla relies primarily on cameras and software, but critics say autonomous cars need hardware like LiDAR and radar to work successfully.

Radar uses radio waves to detect other objects, vehicles, and people. LiDAR does the same thing using light. But Musk, like his cars, doesn't see it that way. It's pretty clear you don't like LiDAR. LiDAR is lame. LiDAR is lame. It's expensive and unnecessary. They're all going to dump LiDAR, is my prediction. Mark my words. When they first launched the system, they had...

eight cameras and a single radar sensor on the front of the car and some ultrasonic sensors. The ultrasonic sensors are very short range, very low resolution sensors that are not designed for automated driving. They're designed for parking assist, you know, to let you know if you're too close to something when you're parking the car.

And here's the thing about a camera-based system: the conditions have to be just right for it to work. If you're driving into early morning sunrise, they can easily be blinded. In the dark, they don't work very well. In fog and in precipitation, they don't work very well. Despite all this, Tesla maintains that its cars have everything they need to reach full autonomy as is. They are creating the impression in those customers that are paying that, that the car is capable of that today.

And it's not. And it very likely never will be with the kind of hardware configuration that they have today. Tesla is marketing and trying to sell you a product that doesn't exist. Nabila Hussain is a lawyer leading one of several ongoing lawsuits against Tesla.

We allege that at no point during the marketing and sale of these vehicles was Tesla even remotely close to having a vehicle that was going to be fully self-driving. And that's true today. She says a big part of all this is the name. If you call something full self-driving, you are clearly suggesting to customers that this thing can fully drive itself.

based on the most recent statements that we've seen, both from Tesla and the regulators, we don't believe that the car is going to be capable of doing that anytime in the near future. And so the plaintiffs here feel as though they paid thousands of dollars to purchase a feature on a vehicle that is never going to

come to fruition in the way that they have been promised it will. What Elon tells the public, what Tesla tells the public versus what they tell regulators is distinctly different. You know, one day he'll go on Twitter or he'll get on a market call and he'll say, by the end of the year, the cars will be able to fully self-driving, won't need a driver. The car will be able to go from San Francisco to New York all by itself. You'll be able to summon the car from its parking spot to where you're standing. And then privately, they'll tell regulators that they're nowhere near that, that you'll need a driver, that drivers need to pay full...

Even the feds are getting sick of Tesla's claims. The Department of Justice is investigating the company's self-driving promises as part of a criminal probe. This is, of course, on top of the lawsuits, multiple recalls, and the hundreds of crashes.

So if the tech isn't ready, why is Tesla doing this? Why beta test on public roads with incomplete hardware, all in pursuit of a product that is supposed to make the roads safer? Maybe because this technology is, according to Musk, the key to Tesla's future. And it's not going well. I do an evaluation of the top companies in the sector. Out of 16 companies I had on the list this year, they were 16th. When we come back, the arms race for full autonomy is underway.

Support for Land of the Giants comes from Quince. The summer is not quite over yet, but shifting your wardrobe to the colder months could start now, little by little. You can update your closet without breaking the bank with Quince. They offer a variety of timeless, high-quality items. Quince has cashmere sweaters from $50, pants for every occasion, and washable silk tops. And they're also available in a variety of colors.

And it's not just clothes. They have premium luggage options and high-quality bedding, too. Quince's luxury essentials are all priced 50% to 80% less than similar brands. I've checked out Quince for myself, picking up a hand-woven Italian leather clutch for my mom. As soon as she saw it, she commented on how soft and pretty the leather was and told me it was the perfect size for everything she needs to carry out on a quick shopping trip.

Make switching seasons a breeze with Quince's high-quality closet essentials. Go to quince.com slash giants for free shipping on your order and 365-day returns. That's q-u-i-n-c-e dot com slash giants to get free shipping and 365-day returns. quince.com slash giants

If you're wondering why Elon Musk is so bullish on autonomous technology, according to him, it's the key to the company's financial future. Companies are evaluated on their future potential. And right now, Tesla is telling investors that the future is one word. Robotaxis.

Kind of sounds like two words, but it's one anyways. Robo taxis. Expect us to operate like a combination of maybe the Uber and Airbnb model. So if you own the car, you can add or subtract it to the Tesla network. You will just summon the car from the parking lot, get in and go for a drive.

Meaning that if you own a Tesla with FSD, in theory, you could eventually be driven to work by your car, get dropped off, and while you're hanging at your office, your Tesla could transform into a self-driving taxi service, a robo-taxi. A large proportion of Tesla's market valuation is based on this idea that at some point they can flip the switch and turn on their robo-taxi service.

Sambabu Al-Samad says the idea is that while you're not using your car, it's passively making you money. Something like $30,000 a year, according to Musk, which is great work if you can get it.

For every ride, those owners would be generating revenue. And of course, Tesla is going to take a significant chunk of that revenue. That's the reason why Tesla is valued by the financial markets so much more than every other automaker because of this idea that they're suddenly going to generate all this pure profit from ride hailing. We expect to have the first operating robo taxis next year with no one in them.

I feel very confident predicting autonomous robotaxis for Tesla next year. Musk said this in 2019. It hasn't happened yet. But the competition isn't waiting around. In fact, each year, Abu al-Samad evaluates the top tech and car companies in the autonomous sector. Of the 16 companies actively pursuing full autonomy, he ranked Tesla dead last. And they have been consistently last every year.

because their system is just not likely to ever work the way they claim. That's because of the whole LIDAR, radar, and sensor thing we mentioned earlier. This year, the top four companies that I had were Mobileye,

Waymo, Baidu, and Cruise. If some of those names don't ring a bell to you, that's fine. They're technology companies. None of them actually make a car consumers can buy yet. But these companies do make cars that you can ride in right now as robo taxis. That's right. Autonomous car companies like Waymo, a subsidiary of Google, and Cruise, which is majority owned by General Motors, are already operating robo taxi pilot programs in a few states like Arizona, Texas, and California.

So how far ahead of Tesla's FSD are these companies? Naturally, my producer Zach and I had to go see for ourselves. So we can just hit start ride. Yeah, there's a screen in front of us in the back seat. We just hit the button and... We went to San Francisco where these cars are operating right now as autonomous robo-taxis.

By the way, these robo taxis are fully autonomous, meaning no one is behind the wheel. It's just us alone in the backseat of a car. We started with Waymo. We'll do all the driving, so please don't touch the steering wheel or pedals during your ride. Finally, relax and thank you for riding with us. We're sitting here in this Jaguar I-Pace from Waymo. Some very pleasant music playing in the background. It's like spa music.

How you feeling? I actually think they do a pretty good job of explaining what this is and kind of putting you at ease. And we sort of move forward. It made a right turn. We're at a red light now. And it all seems fine so far. It's very, very, very smooth. As the car drives around, we see people staring, sometimes taking pictures. These cars tend to stick out because they're covered with sensors, cameras, and spinning LiDAR sets. Got to point this out. We just hit two speed bumps.

and the car slowed down to take both speed bumps at a reasonable pace. And I think that's notable because the Tesla went wild on the speed bumps. There we go, third speed bump. Seamless.

Yeah, that guy just took that turn a little wide. Yeah, and our car was like, yep, nope, nope, nope. And it just kind of slid over to give him some room with no drama. Seamlessly evaded that truck. This is where you start to see a little bit of the promise of autonomous cars. Because that back there was one of those moments like if a human driver, even for a second, hadn't been paying attention, it could have ended badly. But this thing is always paying attention, in theory. It doesn't have a capacity to make the mistakes that we make.

We are in rush hour gridlock San Francisco 5 p.m. traffic. This is textbook, I don't want to be driving right now type traffic because it's so stop and go. It's really boring, monotonous driving. This is the time where you would rather have a robot do it for you and you could just be in the back taking a nap or watching a movie. Yeah, nobody likes doing this.

All in all, Zach and I took five different trips in Waymo and Cruise's autonomous cars. This technology, when it works, is amazing. You can clearly see its potential. We thought Waymo's car drove a little better, but both felt very safe. During another ride in one of the Cruise cars, I looked up and noticed that we were both on our phones and barely paying attention to the road. I forget how adorable a city San Francisco is. Oh yeah, I love it here. It's really great. Well, this is, um...

By the way, we also rented a Tesla to get around on our trip when we weren't testing autonomous cars.

And one day, during a drive back to our hotel, we figured it was only fair to give FSD another try. So I briefly turned it on, and get this, in under five minutes, the car missed two important left turns and then pulled us across a lane headed straight toward oncoming traffic before I had to seize the wheel again. I may have screamed at it when it did this.

So why is it we felt so safe in the cruise and Waymo cars and so unsafe in the Tesla? Well, for one, the robo taxis have all those sensors. Plus, cruise and Waymo are operating in just a few cities, and they only work in limited areas of those cities, areas that have been mapped out extensively. So the cars perform more like locals than tourists. Meanwhile, Teslas have to work everywhere. In other words, Waymo and cruise have more advanced cars with an easier job to do.

And there's a reason these companies have this advantage over Tesla. It comes down to business model. Personally owned cars are a really cost-sensitive commodity. Matt Schwall is the director of safety and incident management at Waymo and a former director at Tesla. Tesla equips all their vehicles with the same sensor suite. And what that means is that they can only equip the cars with the sensors that the most cost-conscious customer needs.

is willing and able to pay for. And the reality is big touchscreens, heated seats, premium sound system, those sell cars better than LiDAR and radar. So the end result is you end up with a camera-only suite. He's saying LiDAR and radar would drive up the price of Tesla's cars, and that would slow down sales. On something that's as cost-sensitive as a personally-owned vehicle, a mass-market car especially, they're expensive.

every dollar counts on a, you know, on a mid-sized car. At Waymo, we're not constrained the same way right now. And that's because instead of selling cars directly to customers, where, uh,

The car really spends most of its time parked in a driveway. Instead, we use our cars for ride sharing. So that means that our cars can add value seven days a week around the clock. I think the difference is in the strategy of how to get there. Kyle Vaught is the CEO of Cruise. You know, we're following the tried and true technique of build a technology. It's initially more expensive. And then once it works, you drive down the cost and you optimize it.

To put it plainly, Waymo and Cruise don't have the same constraints as Tesla because they aren't in the business of selling cars. Not yet, anyway. Right now, they don't even need to make money. They have enough backing from their parent companies. I think the approach that Tesla and others have taken is that start with these really challenging constraints. Like, I have to keep the cost so low that you'll buy it, you know, from a dealership. And I have to strip out all the sensors that would actually make this viable and hope that

then in the next several years or decades, I can make that software fill in all the gaps by that missing hardware. Or I wait for the hardware to be cheap enough that you can close those gaps that way. So that's a bit of a moonshot. And it's the reason you can have Tesla that's fun to drive, but you can't be picked up from one of those with no one behind the wheel.

So while Tesla prays for a technological miracle, Waymo and Cruise look to pilot autonomous robo-taxis in more cities, typically warmer cities. These cars still struggle in bad weather. Right now, robo-taxis are operating in a limited capacity in Phoenix, Austin, and San Francisco.

But to scale beyond this, these companies will have to convince regulators that this is a good idea, which can be a bit tricky because these cars are starting to get a bit of a reputation for just stopping in the middle of the street. Last night's traffic jam came just days after a California Public Utilities Commission vote to expand the use of self-driving cars in the city.

There's a lot of back and forth on this, especially in San Francisco, where residents are getting increasingly impatient with these cars. I'm all about technology, but this is not ready, guys. Jesus Christ, here we are again. When I'm stuck in traffic at 11 p.m. because there's 25 autonomous cars learning how to drive, it's really annoying. The future of cities is not cars. It's only a matter of time until an occupant is put at extreme risk by one of their cars.

So a lot of skepticism on robo-taxis. And this is all happening while Tesla racks up even worse headlines. The rising number of crashes and deaths is not helping consumer confidence, especially because it's easy to conflate Tesla's partial autonomy with the fully autonomous robo-taxis.

At this moment, it's sort of up to Cruise and Waymo to really prove that autonomous cars are safer than human drivers. Both companies have a clean track record with no deaths or major injuries, for now. But Cruise's CEO says he doesn't like to brag about that. We recognize eventually there will be a severe incident, unfortunate moment. And I think when that happens, as a society, we will have a choice on how to respond.

You know, do we let the sort of horrible realities of a tragic incident swing our sentiment towards this technology as a whole, as a category? Or do we, you know, continue pressing forward and maybe raising the bars on regulations or whatever it may be and giving it healthy scrutiny? Americans typically have a lot of patience for preventable deaths caused by other humans. See firearms and, yes, car accidents.

But we have much less tolerance for death caused by robots. I'm worried that as a society, if we're not thinking about that now and talking about the safety benefits of this technology, that we might get that wrong. And that could cause hundreds of thousands of lives. Nevertheless, for all these companies pursuing autonomy, the endgame remains the same. Which is, we all want this technology to be in every car and work everywhere.

Work everywhere. That's the ultimate goal. Drop a car anywhere on Earth, take a nap in the backseat, and it'll take you where you need to go, more safely than a human driver too. The hard truth is that for now, none of these companies are there. For Waymo and Cruise, the technology appears to be much further ahead than Tesla, but the business model isn't proven yet.

Tesla, on the other hand, knows how to sell cars. But its outright stubbornness on hardware and recklessness with how it deploys FSD has things feeling pretty far off course. Despite what Elon Musk keeps telling us. You can drive autonomously with greater safety than a person right now. No one needs to pay attention, meaning you could go to sleep. I really would consider autonomous driving to be basically a solved problem.

Next time on the Tesla Shockwave, the company's not-so-secret weapon. The hearts and lungs of the Tesla growth story are in China. China has now become the world's largest exporter of cars for the very first time. And a significant number of those are EVs. Tesla is helping China dominate the EV race. Can the U.S. catch up?

This episode included clips from ABC 7 News, NBC News, ABC Action News, Inside Edition, Kron 4, KTLA, and AI Addict. Land of the Giants: The Tesla Shockwave is produced by the Vox Media Podcast Network in partnership with our colleagues at The Verge. This episode was produced by Zach Mack. Zach Mack is also our showrunner and senior producer.

Charlotte Silver is our producer. Jolie Myers is our editor. Sarah Craig is our fact checker. Brandon McFarlane mixed and scored this episode. Andy Hawkins is transportation editor at The Verge. Nishat Khurwa is our executive producer. Tamara Warren is our co-host. And I'm Patrick George. If you like this episode, give us a follow and tell a friend. Our final episode comes out next Wednesday.