Michael O’Shea of MOTER

0:00 Introduction to MOTER Technologies and CES 2026

Today in The Garage at an episode recorded live here at CES2026, our guest is Michael O’Shea. Michael is Chief Technology Officer and Chief Operating Officer at MOTER, a usage-based insurance leader. In today’s conversation, we talk about what MOTER’s technology is, how it works by being in the vehicle, providing benefits to drivers and OEMs alike. We talk about the shift to AI in the automotive industry and how our collaboration with MOTER enables easier deployment into vehicles. It’s a very exciting conversation. Let’s go.

0:45 Meet Michael O’Shea

Welcome to The Garage. I’m John Heinlein, Chief Marketing Officer with Sonatus. Today in our in the podcast, we have a guest from MOTER, a usage based insurance leader. I’m happy to welcome Michael O’Shea to the podcast. Michael, so glad you could be with us today.

Why don’t you begin by telling us about you?

Yes, John. Yeah. Thank you very much for including us in the in The Garage. So my name is Michael O’Shea. I’m COO, CTO of MOTER Technologies.

1:10 Michael’s Background in Automotive Industry

I’m a thirty year plus veteran of the automotive industry. I’ve spent most of my career on automotive software with a particular focus on connected vehicles and softwaredefined vehicles. I started my career working for Alpine Electronics way back in the early 1990s and developed some audio systems and eventually navigation systems for that company, including the first line fit navigation system in the US with Honda. It’s still a partner of mine many years later.

After that, I moved to the US and worked for a company then known as Navigation Technologies. Today, they’re called HERE, a mapping company. And there, I worked with automotive ecosystem partners, OEMs, and suppliers, developing location based services and solutions.

And then I started my own company. For almost twenty years, I had a company called Abalta Technologies that provided software services and software products to the automotive industry.

I sold that back in 2018. Just prior to joining MOTER, I worked at AWS with a focus on connected mobility and software defined vehicles. And then I found myself working for MOTER Technologies, a very interesting company at the intersection of insurance and automotive.

2:30 Fun Facts and Personal Stories

And we’ll get to MOTER in a second. You got to tell us a fun fact about you.

Okay. Fun fact. So as you might tell from my accent, I grew up in Ireland.

And upon graduating from college, I moved to Japan.

As you do.

As you do. And it was an interesting time, 1991. There weren’t a lot of people moving to Japan at that time. But in Ireland, we had a government program that was placing graduate engineers various companies in Japan over about a four or five-year period.

And I was one of several hundred Irishmen that Irishmen and women that ended up in Japan. And there’s some of them still living there. And actually, quite a few of them are in the automotive industry. So if you hear those Irish accents, it’s probably coming from that program.

It’s an incredible story. So my fun fact back to you is I lived in Japan for about six months, about twenty years ago. And and it was a wonderful experience. I had been working with Japan for many, many, many years before that, and I had an opportunity to do an assignment there for a while. And I’ve worked with Japanese customers for fifteen, twenty years.

And it’s a wonderful experience. And there’s a million Irish pubs in Tokyo. That is right. As I’m sure, know, maybe that’s why.

Now I understand why.

This is why.

3:42 Overview of MOTER Technologies

So it’s it’s wonderful to have you. So tell us about MOTER for our listeners who may not know about it.

Yeah.

So MOTER Technologies By way, want to say MOTER is spelled M-O-T-E-R because everyone of course thinks of “motor”.

Right.

And that’s not a misspelling, actually. It’s Mobility on the Edge in Real Time. And I’ll explain why we call it that as we go. But essentially, a company sitting, as I said, between automotive and insurance companies.

It was a company spun out of a very large Japanese insurance company, Aoi Nissay Dowa Insurance, which is part of the MS&AD Insurance Group, so a significant backer there. The key insight that led to the formation of MOTER is that modern vehicles are equipped with lots of sensors, cameras, radars, LiDARs even. And the data coming off of these sensors can be extremely valuable to insurance companies. Traditionally, they’ve relied on OBD, like dongle-type devices or smartphones, to gather information about acceleration, braking, cornering, and speeding, ABCs.

So these are kind of proxies for risk. But if we understand the context, we can do a much better job of really measuring risk and fairly attributing risk.

For example, if you brake aggressively because somebody runs out in front of you, you shouldn’t be penalized for that, right? So that was the idea. And what our intention is is to get our software running on the vehicle where we can access that data locally, keep the data on the vehicle for privacy reasons, and limit the amount of data we’re sending to the back end to reduce cost. So hence, mobility on the edge in real time. We want to be on the edge. We do some cloud integrations, but primarily it’s integrated.

5:27 Innovative Approaches to Usage-Based Insurance

That’s okay. So then let’s double click on that. So tell us about how your approach with deploying UBI or these driver scoring in the vehicle differs from conventional approaches. Let’s go to the next level detail then.

Right. So I think the conventional approach, historically, it started twenty-plus years ago with dongle devices that are plugged into the car. It’s not very reliable. Different vehicles can draw up different issues.

And sometimes customers will play with this. They’ll unplug if they’re intending to drive fast to CES from Los Angeles or something like that. So it’s not the most reliable way to gather data. And it’s also expensive.

There’s an operational challenge in getting dongles out to customers, collecting them after the fact, and so on. So along comes the smartphone. And the insurance companies said, Okay, these are everywhere, maybe we can build a usage based insurance application on top of a smartphone.

And indeed, many of them do so today. But again, lots of issues. The accuracy of the data, the ability for the user to change the permissions, and so on, such that you’re not getting a good score.

All of these…knowing who’s driving, for example.

And knowing who’s driving is a critical issue as well. So there’s some limitations with these. And of course, they’re not accessing all that rich data that is on the car that gives you the context to really do this in a manner that gives you greater fidelity on the risk score and fairness for the driver.

6:58 Addressing Privacy Concerns in Data Usage

Well, and as you’re doing that, monitoring, if you will, monitoring and looking at driver behavior, privacy must be a concern. So what are some ways that you can ameliorate that and address privacy concerns?

Yeah, privacy is really top of mind for us, and obviously for our OEM customers and their customers, the drivers.

A lot of issues in that area. There’s been some controversies, especially in terms of how data is used by insurance companies. So no customer wants to find that their data is finding its way to an insurance company, and it’s impacting their rates without their knowledge and consent. So for us, consent is critical.

Everything we do is with customer consent. We’re very transparent about it. It can be withdrawn at any time. The data can be deleted if they choose to do so.

So that’s built in from day one. But it’s also privacy by design. And our edge-focused approach helps with the data privacy issue because simply we’re not pulling the data off the vehicle to begin with. What we’re doing is computing scores on the vehicle and with the customer’s consent, sending them to the back end to the insurer to give them hopefully a better price on insurance.

That’s great. And we have, Sonatus has the same approach to privacy. We care about it a lot. Our OEM customers, in fact, are leveraging — in a similar way to what you’re doing — they’re leveraging the configurability of our products to ensure that they’re respecting the opt in opt out preferences of our drivers and customers. So I think it’s fantastic approach. So it’s clear that drivers can get a benefit from having a better scoring, assuming they’re a safe driver, of course, from this approach. But what are some benefits that OEMs and other people that value chain get as well?

8:37 Benefits for OEMs and the Automotive Ecosystem

Yeah, especially this is one of the rare occasions where everybody benefits if it’s done right. So for OEMs, they’re all very committed to zero fatalities by encouraging drivers to be safer. We can lead them further down that path.

But there’s also many opportunities for monetization here. So we provide custom insurance products for our OEM partners. An example would be Afeela, the Sonatus Honda Mobility company. We’re the Afeela insurance company, white labeled for them.

And in cases of companies like that, they can offer custom insurance products, and they can participate in some of the revenue flows there. Insurance companies are very interested in attracting quality, low-risk drivers. They’re interested in claims data, they will pay for this. So we share revenue with OEMs.

Again, assuming the consumer consents, we will share the revenue from those data insights that we sell to the insurance companies.

It’s fantastic. And I think we’re always looking in my career really, lot of my work has been around partnerships and finding win-wins. So it’s always exciting when you can see a place where there’s multiple layers of benefits. So you’re all marching down the field together and very excited for everyone to win.

9:58 Enhancing Brand Loyalty through Technology

Yeah. And it goes for the OEM well beyond that sort of direct insight monetization. We bring a lot of value back into the ecosystem. And people wonder how’s that possible?

Well, we create custom endorsements. If you have an issue with your car and you need parts replaced, we will bring the customer back to the dealership by guaranteeing OEM replacement parts. If you total a vehicle, we might waive the deductible if you buy an OEM-replacement vehicle. So there’s a lot of additional value that comes into the ecosystem for the OEM.

It’s actually very significant if you add all of this up.

Yeah, we’re showing a number of different demonstrations here at the show this week where technologies and the right kind of technology collaborations can build brand loyalty, whether it’s upsells or brand loyalty or getting people to branded service centers in various different ways. And I think it’s actually an often, I think, underappreciated benefit of software-defined vehicles and the kinds of things we’re doing is that you can actually have more affinity, not less affinity, to the OEM if you do it right.

11:02 Collaboration with Sonatus and AI Director

I totally agree. Know, a big issue for a lot of our OEM partners really is the total cost of ownership. So there’s a purchase price of the vehicle, but you’ve got to consider insurance. And oftentimes, insurance is the next most expensive item. So if we can help lower that cost, the consumer appreciates that. And we can do that through the kinds of programs we’re talking about today.

Well, we have to talk about our collaboration together. You’ve been a wonderful partner with us. And just outside the podcast booth here, we can see our joint demonstration here at CES where we’re showing your model deployed onto an ECU using our Sonatus AI Director infrastructure. AI Director in brief is means that third party AI models of all types and sizes can be deployed into the vehicle, not in the edge, not in the cloud, but into the vehicle. And but I’d love to hear what your experience was and how did this infrastructure help you in our collaboration.

Yes. That’s a that’s a very interesting story. So for MOTER, fundamentally, what we’re doing is generating models that take this data that I’ve talked about and generates a score that essentially predicts the likelihood of a future collision. Now, those models are very expensive and complicated to develop.

They’re developed by data scientists and actuarial scientists.

And they have to be approved by government regulators in every state that we operate in.

Oh, is that right?

Yeah. It’s really quite a complicated process. So that’s just the first step. So we have the models.

They’ve been approved by all the states. Now insurance companies can use these for underwriting purposes. But the next step is we’ve to get them deployed onto the vehicle. And that’s where we hit challenges.

If we work OEM by OEM, everything is bespoke, It’s a very long and very slow process for us. It’s also hard sometimes to get access to all of the rich data that we need. So AI Director was a wonderful discovery for us, where we found that we could deploy our models through AI Director in a standard way with a consistent interface to data with the compute requirements that we needed, which are not a lot, but we certainly need some compute.

So a very easily deployable mechanism to get our software on the Vehicle Edge. So this is great for companies like us.

Yeah, a couple of points to match that is one of the things that we’re so excited about and we’re showing here in our show is that the infrastructure we’re providing, that an OEM if an OEM commits to putting that infrastructure in, they can then deploy each incremental model like yours and others very easily versus if they had to start from zero on every individual model that came in, the integration costs can be high. The second is you mentioned data, and data is a sort of lifeblood of Sonatus. We we come from a data center background and expertise. But one of the things we’re doing is providing this this framework layer that allows us to sort of serve to you, provide to you. data from across different ECUs across the vehicle. So in the same way, you don’t have to bespoke integrate with the ADAS system and the driver monitoring system and the sonar system or whatever. You can benefit from those in a more reusable way.

14:12 MOTER-Sonatus Case study and Demonstration Insights

Yeah, this is a huge benefit for companies like us. We also don’t want to have to deal with ASIL, strict ASIL requirements on an ADAS domain controller or something like this. If we can find a central place to run where we get access to that data, and can do it in a compliant way, that’s wonderful for us.

So you also did a recent case study with Sonatus showing the benefits and how this can be easily deployable. We’ll put that in the show notes so that the listeners can click and take a look at that. And we’ll also include a clip of the video, the demonstration showing how this this demonstration works. It’s you mentioned it’s very lightweight model. But I think a lot of times the misconception is AI in the vehicle is limited to kind of autonomous driving or ADAS, which can be quite heavy compute. We’re talking about a few percent of a single Arm CPU from many different vendors is all it takes to run this model, which is de minimis. And we’re proud with the Director, we’re able to not only to provide those kind of bowling lane limits, so the model has to live within a specific memory footprint, within a specific CPU footprint.

So that’s another reassurance that it’s not going to impact other compute happening side by side with it on the same ECUs.

That’s absolutely correct. And I think, you know, the common misconception that we run into, because we rely heavily on computer vision, is that we’re running some kind of computer vision process within our model. We’re not actually doing that. What we’re doing is we’re taking the outputs of these computer vision systems that are already built into the car. So we’re looking for events that are coming from those systems. And then we’re doing some calculations. So actually a very lightweight model, but we can drive enormous benefit from the processing that’s already happening in the vehicle.

That’s such an important observation. And I’ve been explaining that to everyone at this show as they come by is we’re not reinventing the wheel. The ADAS system already knows if you’re following too closely because it has to pretty much by law be prepared for automatic emergency braking, for example.

So using that data that’s already there but providing it to your model means that the work you have to do is much less but yet you create value.

Exactly.

So you’ve been deployed. You mentioned Sony Afeela. What other places you’ve been deployed?

So we work very closely with Toyota, and there’ll be some more news about that in this year. Sony-Honda Mobility in Afeela…. And there’s another OEM that we’ve signed with that we’ll be announcing in the next few weeks and more to come. So quite a lot going on. We’re also here at CES showing our new DriveSage™ coaching application.

16:45 Introducing Drive Sage Coaching Application

It’s one thing to measure your risk and your score, but we’re really committed to providing information back to the driver so that they can understand what behavior is leading to risk, which is leading to a bad score, which is leading to a high insurance price. So if we can get that information to them in a fun and exciting way, then we can really hopefully influence behavior. So that’s what DriveSage is, and it’s an additional offer that we make through the Android Automotive OS app stores.

It’s really smart. I drive an EV. I’ve been driving EV for twelve, fifteen years now. And if you turn on the monitor that shows you the energy consumption of your driving, you very quickly realize, oh gosh, I guess I probably could have accelerated slower and saved energy. So in a similar way, the ability to feedback to a driver who, let’s say, wants to drive safe, maybe not everybody wants to drive safe. But imagine your customers probably want to drive safe. The ability to help them understand how to do that better is a win win for everyone.

17:41 Engaging Younger Drivers with Data Sharing

Absolutely. And I think what people sort of tend to forget with regards to being monitored and that a lot of people are initially skeptical. And most of those people honestly are my age, right? They come along and they say, I wouldn’t do this. But think about younger drivers.

Their insurance costs are extremely high.

Very high.

And they’re a lot more accustomed to sharing data if they get some value in exchange. As long as it’s been done in a very open, transparent, and controlled way, they will do it. And it’s a great benefit to them. And we all benefit if they’re safer on the road too.

18:13 Building Trust through Transparency

Absolutely. Transparency is key. I think everything that we as an industry do, we’ve talked about this many times on the show, everything we do if we’re transparent about it is going to build trust. And it’s going to build adoption if people realize that they’re getting a benefit and they know what you’re doing with it. Exactly. So you’re walking around the show, I know you’re meeting with practically everyone. What are some trends you see or what’s on your mind this week?

Yeah. It’s I have to admit, I spent most of it standing at our our partner’s booth, Yazaki, down just a few few booths down. So I didn’t get a lot of time to walk around. But it’s pretty evident that this year, it’s AI, agentic AI, it’s robotics.

AI is everywhere. We’re part of that, you’re part of that. So I think we’re quite on trend.

18:56 Closing Thoughts and Future Conversations

Well, it’s wonderful to speak with you. We’ve known each other for a long time. And every time I see you, I’m always happy to chat with you and you always have insightful points. I’m so glad you could share it with the audience. Thanks for coming.

Yeah. Thank you.

If you like what you’re seeing in this episode, please like and subscribe to see more like it. We’re having episodes here at CES as well as around the world at various shows and back at our home studio. We look forward to seeing you in another episode of the podcast very soon.