Listen: Next in Tech | Episode 66: Connected vehicles in transition

View Full Transcript

Transcript provided by Kensho.

Eric Hanselman

Welcome to Next in Tech, an S&P Global Market Intelligence podcast where the world of emerging tech lives. I'm your host, Eric Hanselman, Principal Research Analyst for the 451 Research arm of S&P Global Market Intelligence. And today, we will be discussing connected vehicles with 2 returning guests: analyst Mark Fontecchio; and senior analyst Johan Vermij. Gentlemen welcome back to the podcast.

Johan Vermij

Thank you, Eric. It's good to be back here.

Mark Fontecchio

Yes. Thank you, Eric.

Eric Hanselman

So we've talked a lot about the various aspects of the Internet of Things. We've talked about some vehicle perspectives. What I wanted to do is really sort of get into the connected vehicle itself and a lot of the different aspects. And it really is that sort of the culmination of all the different pieces of managing. We talked a little bit about autonomy, all the different aspects that come together.

But really, it is that epitome of digitized environments, things that are generating a lot of data, consuming a lot of data and a lot of the interactions that start to make a lot of the different actual implementation details so interesting and complex. And Mark, there's a Voice of the Connected User Landscape study that's out recently that's got some perspectives on this. So what's the study say?

Mark Fontecchio

Yes. So in terms of the study and the survey, I think it's kind of important to set the ground rules on what the survey covered. So what do we mean by connected vehicles, and then what do we mean by connected vehicle services, which is what the survey sort of entailed.

So for this particular survey, it was fielded in January, we asked about connected vehicle services from automakers, which we often call OEMs or original equipment manufacturers. And so really, the most familiar one that listeners will probably know about is GM OnStar. But all of the OEMs or nearly all of the OEMs have some version of that.

So Toyota has Entune, Chevrolet has MyLink, BMW has ConnectedDrive, kind of goes on and on. And so these OEMs, they're trying to be part of what we might call the subscription economy. We all seem to have a lot more subscriptions than we used to: streaming media subscriptions, meal kit services, monthly shaving clubs, you name it. And so the OEMs kind of figured, hey, why not us? Why can't we get consumers to pay for another subscription?

Eric Hanselman

There's a revenue stream in there. So why don't we tap it?

Mark Fontecchio

That's right. So what we're seeing in this study data about the adoption of connected vehicle services, I think the top line takeaway is that the adoption growth thus far is extremely slow, if not flat. So when we surveyed consumers last year on this topic, 27% said that they had connected vehicle services. This year, that number inched up to just 29%. So that's pretty much within the margin of error. So I think that's the main takeaway from the survey, and we can kind of build off that in our discussion.

Eric Hanselman

Well, it raises that interesting question, which is that piece of we're happy to consume a lot of subscription services when they've got what's primarily an entertainment focus. But we start thinking about the transition to what that looks like for the vehicles we drive, that's a slightly different kinds of -- kind of thought. And we think about vehicles generally, it seems like that hasn't been something where we've expected that there is some sort of monthly tie that's specifically tied to the vehicle.

Now of course, we do this for communications. We got cell phones. We've got all these different things that we subscribe to where we've got value that's really abstracted into the communications or the video or the music or all the entertainment pieces that are there. But it seems like looking at cars that it hasn't extended in quite that same way. Any ideas from the survey what attitudes are about that? Or is it something that's sort of beyond that typical thinking?

Mark Fontecchio

Yes, that's a really interesting point. And I think a lot of it just comes down to consumer expectations. I mean, many years ago, if you think about media in general, many years ago, people did not pay for watching media on their television. They had the rabbit ears on top of their television, and they got the channels that they -- specific number of channels that they got, and they got it for free and then kind of moved into cable subscriptions.

And so consumers have slowly kind of gotten used to paying that monthly subscription fee of some kind for television services that they want. And so that's kind of made it a little bit easier for those streaming media companies like, say, a Netflix or a Hulu or what have you to charge that service, whereas something with vehicles and connected vehicle services, consumers just aren't used to paying for that in any way.

Like you said, they're used to paying for a smartphone subscription or they're used to paying maybe even for a streaming media service like a Spotify or something like that, but they're not used to paying connected vehicle services. And so one of the other takeaways from this survey is that these connected vehicle services, they offer a lot of things that consumers can already get on their smartphones.

So they're already paying that sort of data plan, that cellular data plan, and they're getting a lot of the apps that they might use in the car, so navigation or traffic and road condition alerts or streaming media services like Spotify or what have you. And so they're not going to really be willing to pay extra to get that in a connected vehicle package when they've already got it on their smartphone that they can bring into their car and connect to Android Auto or Apple CarPlay.

Eric Hanselman

And they've already got a way to be able to handle that. I guess the question is when we start thinking about that, are there other aspects that they could potentially expand into? And, I guess, Johan, if we think about some of the pieces that we've typically looked at in terms of other delivered connected services, and there, right now, what you get for a lot of those connected services are fairly narrow, to Mark's point.

You're really looking at maybe some emergency services, maybe some nav services, but there are other ways to generate them and ways that are more comfortable for other users. We start thinking about some of those next stages. I mean, we see bits and pieces of this when we look at Tesla's services and capabilities. There, you're talking about over-the-air updates to the car itself.

We're still sort of in that transitional period for other manufacturers. Are there aspects of that? Are there other places that we could potentially go when we start thinking about those capabilities?

Mark Fontecchio

So yes, Eric, we did ask consumers what connected vehicle services they think are worth paying for. That was a new question that we asked this year. So as you can imagine, several of the top results were vehicle-specific. So if you think of OnStar, its biggest selling point, when you see an ad for them is that it pushes its emergency services where your car breaks down or you get into an accident, you can contact someone immediately.

And so those were the top results as part of the survey. Roadside assistance was the #1 feature that consumers thought was worth paying for. Stolen vehicle locator was another. So I mean, many years ago, you think about LoJack, right, and the thing you would lock on your steering wheel. So that's a service that people are willing to pay for. Remote lock and unlock was another. So you kind of get the idea if the OEMs want consumers to sign up for their connected vehicle services, they need to be services that can't be as easily done on a smartphone. And then you kind of take it beyond that.

You're kind of mentioning over-the-air updates that Tesla provides and then some sort of assisted driving features or autonomy features or even sort of customized cockpit features that could be included like, say, you have 2 people in your household that are driving the car. One person gets in. If their smartphone is connected to the vehicle, then the seat lines up in the perfect spot.

And maybe for one household member, the seat warmers come on. And for the other, the temperature gets turned down to 65 with the AC blasting because that's what that person -- so you can see that there's a lot of opportunity and potential for growth and improvement in these connected vehicle services for sure.

Eric Hanselman

Yes. So vehicle-specific things that you can't do with your mobile device. We start getting into those kinds of areas. I mean, Johan, what do you think are those angles that start to have interesting impacts, but there are also aspects that we're starting to get into areas of concerns about security, data privacy, those other pieces as well?

Johan Vermij

Well, we've been driving around dino juice-powered vehicles for almost a century now. And the whole concept of the car, the whole architecture is changing where it was like those little machines powering wheels and individual parts, it's now becoming one giant network with compute capacity. So as it becomes connected, it becomes vulnerable like we see in an office environment where you see the DDoS attacks on the enterprise network or recently now that we're getting into IoT and we're putting sensors in every machine and factory, we see malicious attacks on pipelines and critical infrastructure.

And I think a lot of people are worried about this moving towards their vehicle because I think that's one of the main inhibitors for adoption of autonomous vehicles is that they know they speed with 100 miles per hour or have the feeling that they're in control, especially in Europe, driving stick shift. And what if you have to give that control to an artificial intelligence?

Eric Hanselman

So that raises the question of -- and not only the artificial intelligence but just ownership of the data because as you're identifying, we throw off a lot of data. And there's been -- I don't think we've got numbers on a lot of the onboard data monitoring that's happening with insurance companies, which are you get a favorable rate if you feed us the car performance data.

But clearly, there's a segment of the population that's going to be concerned about disclosing that data. We already know that electronic tolling information has privacy concerns. It's wound up in lawsuits being able to identify a location of who is where and based on electronic tolling.

Maybe what we wind up needing is the manual transmission, the stick shift for how we expose data. And I wonder if users had more control, Johan, to your point, whether or not that would help to allow them to be more comfortable with that sort of data management because it does seem like, Mark, the point you're identifying is that users are concerned about how that data is going to get used and the kinds of things they want to be able to do with it.

Mark Fontecchio

Yes, I can speak to that really briefly, Johan. In our survey, in addition to asking about these connected vehicle services, we actually asked about usage-based insurance. So you were mentioning that earlier, Eric, in your question.

And so there are not very many people who have usage-based insurance right now. So that's basically the concept of -- you have a plug-in device or a smartphone app that monitors your driving behavior. So it makes alerts when you're braking hard or accelerating too hard or turning too quickly, maybe speeding too much. And then your insurance rate can -- if you're a good driver, your insurance rate can be lowered.

And so there's that sort of trade-off with sharing how you're driving with the insurance company in exchange for getting that better rate. And so the majority of the consumers that we surveyed, the vast majority, I think -- I think it was only about 1 in 10 said that they actually have UBI or usage-based insurance.

So the ones that don't, we asked them why they don't. And the most popular response was they're just happy with their current insurance rate. But privacy concerns was up there. 1 in 5 said that privacy concerns was a big reason why they didn't have a UBI smartphone app installed. So that's definitely top of mind for consumers in terms of the sort of data that they're willing to give up.

I think the benefit -- maybe, Johan, you can speak to this, but the benefit really needs to be there in the trade-off. If you're going to share all this driving data, you really need to see some sort of financial benefit or some other type of benefit for doing so.

Johan Vermij

Well, yes, if we look at social media and news online, everybody is willing to give up privacy to get to free content. But I think it's -- the privacy aspect is not yet up front and center for the OEMs in development. I think their focus primarily lies with that security of securing the vehicle and in getting that trust for autonomous driving and preventing AI bias.

The privacy, I think we're now buying our first generation of smart cars that collect this privacy sensitive data. And it will get more important once we get to that point when there will be a big aftermarket in secondhand car sales and spare parts. And if you buy a new secondhand or replace your navigational system and you get the navigational system of an important public person, and you see where he's been for the past 2 years, every week, then -- well, then yes, you will have a concern as an automotive company.

Eric Hanselman

So, I guess, the question that gets to is that you're talking about some relatively high-level concerns about the data privacy pieces and potential data exposure. We went through, what, it's been now at least 5 years since we ran through a lot of the car hacking that was going on, people who were taking over cars, basically hacking in through various car sensors that were poorly protected and being able to steer the car, hit the brakes, disable the brakes, all these other scary scenarios. Do you think people have gotten over that piece? Are they less concerned about those fundamental aspects? Or is that still looming back there?

Johan Vermij

I think the industry has made big progress in that regard. Similar to the general IoT deployments, the first sensors were okay. It's cool what we can do. And now everybody's thinking about security, and we have updated regulations. The industry really has started to take the connected car serious.

Companies have started to address security in a variety of ways such as, for instance, semiconductor vendor Rambus, which is addressing the challenge at the silicon level or [ Segrador Labs ] addressing the root of trust for components through public key provisioning or even companies like NXM Labs providing blockchain-based security. Then there's Karamba Security, which brings its core wall offering and SafeCAN to then to get messages on the canvas level.

And other founders in space include industry vendors like Harman and Continental and some tech companies like Thales and some start-ups like GuardKnox, SafeRide, Upstream Auto and Vector. As far as I can see, and I don't know if that survey data taps into that as well is that the primary concern for consumers not as much the privacy or the car being hacked. But I think the biggest fear is in can I trust the AI to make that right decision.

And I think that's one of the focus points for the automotive industries. Well, if I look at the most recent NIST publication of March 2022 by the National Institute of Standards and Technology, it released a special publication for managing bias in artificial intelligence. It's like, okay, how do we program that AI to make the right decision. And can we trust it? And is it accountable?

For instance, if I -- you have that old trolley conundrum where you have to choose which one dies and which 1 lives. Do you hit the brakes, and do you kill someone that's hitting you in the back and save someone in front of your car or the other way around? How do you make that choice? And how do you program that AI?

And everybody that's programming AI puts in bias to somehow capture in algorithms what do you think is important in life. Do you kill an old man and save the young people? Or in a different society where there's a huge respect for elderly people, do you program it the other way around? Or in the U.S., well, you tend to save a human life in favor of a cow, but in India, maybe it would be the other way around. So how do you manage that AI and have it accountable?

Eric Hanselman

Well, there's a lot that starts to get into that sort of situation when we're dealing with much more complex issues. And right now, we're -- at least today, we're in an environment where sensors aren't going to -- we don't have enough data -- or I guess, fortunately or unfortunately, to make those more nuanced decisions.

But even the fundamental things when -- Johan, to your point, it's -- there are a lot of different aspects of how the car is going to behave in an autonomous environment that really have a lot of concerns and areas that are getting very gray and start to get into societal values in terms of how they actually operate. And, I guess, I wonder, is that the advantage of a connected vehicle is that you could, in fact, dynamically change what's there.

I mean, when you've got Teslas who have got all these joke features for -- I don't know, for those of you who haven't heard, with your Tesla, you can download a whole bunch of different things that are just sort of goofy, not particularly functional aspects that -- to the car itself. It's -- some are functional in which what noise is the car going to make when it's proceeding slowly so that you can let pedestrians know you're coming to playing gags on people who are actually passengers in your car.

There's a lot that's in there that starts to get into some really interesting areas that we haven't traditionally associated with cars. And Mark, I think it gets back to your original point. These are things that they're vehicle-associated aspects that are a big part of this, but maybe the change that's necessary is to start to expand what are those set of expectations.

Mark Fontecchio

Yes. To Johan's point, we didn't -- we don't ask about autonomy in this particular survey, but another survey we do. And I would concur with Johan in terms of comfort levels. I think the comfort level is less about automotive hacking and more about just being out of control of the vehicle.

And so we ask the consumers what kind of autonomy would you be most comfortable with? Or we ask them about different kinds of autonomy. So it might be like a robotaxi or it might be kind of like a shuttle on a college campus or something like that.

Well, they're most comfortable with autonomy when it's like a last-mile delivery robot like one of those ones rolling around a college campus delivering a pizza to a dorm or something. They are plenty comfortable with that level of autonomy. But when it comes to robotaxis driving around a busy city, no, they're not as comfortable with that.

And so -- and I think a lot of it really has to do with perception. If -- I think it was a couple of weeks ago, an autonomous vehicle was pulled over. And that like made the national news that an autonomous vehicle was pulled over. And I think it like pulled up a little ways in order to like get into a more safer area, and so that was the big story.

And so we have how many thousands of fatal accidents on the road every year from human-operated vehicles. But the one that makes the news because the news contains the word new is an autonomous vehicle that actually got pulled over but didn't harm anyone or do anything bad.

And so I think consumers just need to get more comfortable with the idea of sitting in the backseat while the car drives itself and being able to, like Johan said, trust the AI of that vehicle to make decisions that are as good as or really ideally a lot, a lot better than a human ever would. And I think that's what it's going to take for consumers to be comfortable with autonomous vehicles eventually is that they're going to have to be safer an order of magnitude much greater than human-operated vehicles.

Eric Hanselman

Well -- and I'd make the case that it gets back to the thing that we talk about with ML and AI all the time, which is establishing trust. And it's -- Johan, on your point, it's explainability. It's being able to explain to someone why is it going to do what it's going to do to be able to get that level of predictability and build that trust so they can actually do something about it.

I want -- if we're getting to the sort of general to a transition in which this is possible, what should organizations be considering in their plans around connected vehicles? We talked a little bit about insurance. I mean, is this something where market sounds like some of -- if there were incentives to make people move to a usage-based insurance, if your insurance came along with a SiriusXM radio subscription, would that be enough?

Some of that -- we're getting towards incentives that get people to try and shift to that. Should businesses be thinking about integrating with connected vehicle services and capabilities? And where should they consider that heading? And I'll throw that to either one of you first.

Mark Fontecchio

Yes, I can definitely start it off. So what the OEMs are doing now is they're offering free trials, and oftentimes, they're very lengthy. Like if you subscribe to HBO Max or something, you get like a 7-day free trial before they start charging your credit card.

Well, with these connected vehicle services, they're often offering like a year or more of a free trial for the connected vehicle services. Yes. So -- and the number that I mentioned earlier, how 29% of consumers have connected vehicle services, well, a lot of them are -- only about half of that 29% are actually paying for it. The rest of them are on some kind of free trial.

Eric Hanselman

Oh.

Mark Fontecchio

And so now that kind of sounds like not very good news for the OEMs, but there's a bit of a silver lining because those that are on a free trial, about half said they plan to continue the service when the free trial is over and only -- and 1/4 said they weren't sure. So only 1/4 said no, we're not going to continue after the free trial. So that's kind of reason for a little bit of hope, but it's also a question about their intention for the future. And sometimes those intentions don't really turn into reality.

The other thing that I will say that is, I think, is a hurdle to adoption is cost. On average, our survey respondents said they'd be willing to pay about $15 per month for connected vehicle services. The packages that OEMs are offering are typically much higher than that.

I was just looking at OnStar this morning. And one of the most basic ones from OnStar, for example, is $25 per month. And that's just kind of like their very basic package. It goes up from there up to about $50 per month.

So I think there might be some recalibration necessary from the OEM side to realize that consumers aren't going to really want to be dishing out $50 a month for roadside assistance services. So it's either they need to start offering a lot more services or they need to bring that cost down.

Eric Hanselman

Especially the perceptions around what subscription services cost broadly. And when you get entertainment services that are in that $10 to $15 range, that can look like a pretty steep hill decline.

Mark Fontecchio

Yes. I mean, you know how it is with these subscription services. If they're low enough like you don't even -- they happen, and you don't even realize it. It's like a credit card charge, and it's just part of your overall bill. But if it's something that is a little bit higher, then you might take notice of it.

So they need to maybe follow that sort of like Planet Fitness model, where you make that subscription price so low that the consumer is just -- it's too much of a bother to even cancel it. You know what I mean?

Eric Hanselman

So Johan, what are your thoughts? Are there capabilities that might potentially open that door a little more, things that will be more attractive?

Mark Fontecchio

What do you think, Johan? I mean, I kind of mentioned those sort of customized cockpits features. I think if OEMs started offering that a little bit more, then that could be incentive to pay more. I do think that if there is some way for -- like GM OnStar, they have their own insurance division.

So you can get insurance through GM OnStar and conceivably get better rates if you have good driving. And so I think if the discount on the insurance side is enough to offset the cost of the connected vehicle services, that could be enough of an incentive as well. What do you think, Johan?

Johan Vermij

I think there's a lot going on in the automotive industry. It's not just the further digitalization of the car itself and the entertainment system and what have you got, the connected car feature services. But we're also in progress in electrifying our fleet and consumers start connecting their vehicles to home battery storage. And that's a part where the automotive industry also plays a role.

So they're also innovating on that end. And I think if that whole experience becomes more seamless, like you said, with a personalized cockpit that adjusts to your own preferences, but when it's parked at your -- from door that it charges your house or interacts with your battery storage and your solar panels.

I think it's -- if that whole service package extends, and the eventual benefit is not just insurance but also a lower energy bill. And all these parts together are the way forward. And in terms of the privacy concerns, I think the automotive industry should really focus more towards that aftermarket that is still nascent for these types of cars, but it will be the challenge for the next decade.

Eric Hanselman

This is one of those things where -- that the aftermarket add-on enhancement, that's always been a big cash revenue source for the car industry in general. So interesting things there. Although, as you were saying, we do have now a small number of cars that can already be the home battery themselves.

Ford F-150, you've got a handful of small sedans that are doing that now. Maybe that's that entree. Interesting perspectives. Well, these have been really interesting insights. Thanks to you both for being on the podcast. So great having you back.

Mark Fontecchio

Thank you for having us, Eric. Yes, it's been a great chat.

Johan Vermij

Happy to be here as always, Eric.

Eric Hanselman

And that is it for this episode of Next in Tech. Thanks for staying with us. And thanks to our production team. It includes Caroline Wright, Caterina Iacoviello, Ethan Zimman on the marketing and events teams and our studio lead, Kyle Cangialosi.

I hope that you'll join us for our next episode, where we'll be discussing the mergers and acquisitions markets, especially looking at security with Brenon Daly and Scott Crawford. I hope you'll join us then because there's always something Next in Tech.