Upstream & Infosys Discuss AI in Automotive

Transcript

– Welcome to Upstream’s tech talk series. I’m Fay, from Upstream Security, and we offer the first cloud based cybersecurity solution that’s purpose built to connect protected vehicles and smart mobility services from cyber threats and misuse through the use of data. I’ll be the host of this Upstream and handing it over to you to introduce yourself.

– Good afternoon, Fay. Great to talk to you. My name is Shridhar, I’m based out of Dallas, Texas. Have been with Infosys for about 20 years now. Of that, I have spent about 17 years in the automotive industry. I have great passion for cars, I love the automotive industry and the work that we do. And I have also been very fortunate to work with almost every automotive OEM, the top ones in this, you know, in this, in this space, great to talk to you.

– Pleasure, pleasure. Can you tell me a little bit about what Infosys does to serve viewers or become a little bit aware of that before we get into the depth of our conversation?

– Yeah, Infosys is a technology consulting company. You know, we are about or 240,000 employees, globally. Work in most of the countries have a very, very large global presence. What we do with respect to is, as I said, technology consulting, we are spread across multiple verticals. I am part of the manufacturing vertical, but we do work with financial services, banking, you know, communications, media, entertainment, service. utilities, and you name it, it’s there. But primarily what we do is we help our clients in terms of multiple digital transformation, making sure that your existing, you know, systems are up and running in the most efficient way. And we do a lot of business consulting as well with our clients. Cybersecurity in fact is one of the upcoming and growing practices for Infosys. And we are happy to see a significant value that we are adding in the cybersecurity space with our clients.

– Great, you mentioned that digital, digital transformation now within that digital transformation, we have that the rise of the connected vehicle of the shift of the automotive space to a digital space. Now with that rise of connected vehicles, and that ecosystem becomes a rise of automotive data. How do you see understanding that data, whether that’s through AI or other methods playing a role within that connected vehicle space?

– Yeah, that’s an interesting question Fay. You know, if you look at automotive companies over the last many years, there is a significant amount of that data that’s getting generated. And I’m going to take a step back and, you know, kind of give this particular answer. There’s a lot of that is being done by the automotive companies in the space of AI, with respect to autonomous vehicles, right? I think that’s a given. They have formed different subsidiaries. They have formed different companies, and there’s a significant investment in that space. And if you really look at AI, AI plays a great role in the autonomous part of the vehicle itself, so you know the manufacturers are developing. However I’m going to not talk about that, because that is a topic by itself and a subject by itself. But if you really look at it 2019, there were one 91 million vehicles that were produced or manufactured by the auto makers globally. And that’s a significant volume of vehicles. And what does that entail for an automaker, right? If you look at automaker, automaker has broadly, I divide into two different functions that they deal with. One is the enterprise functions. And when you look at enterprise functions, they are like finance, legal, procurement, HR, which I wouldn’t say is very different for any large organization, right? But I, I primarily focus and, you know, I want to focus actually on the core functions of an automaker or a manufacturer, because that is where it is very significantly different as compared to any other industry segments. And if you really look at, you know, the core functions of an automaker or a manufacturer, it starts all the way from planning, conceptualizing, designing, manufacturing, sales, and marketing, and after sales, right? These are the core functions. The role of AI in enterprise functions is significant, but at the same time, I think AI is kind of adding a significant value, and kind of making a deep impact on the core functions itself. If you really look at, for example, starting from planning and conceptualization, that will have a significant amount of data that goes into the analysis of forecasting and AI plays a significant role in the forecasting itself. And forecasting is not just based on the historical data, but you combine a bunch of in-and-out all data to kind of, you know, make meaning out of it. Right? And then manufacturing again is a large topic by itself. There’s a lot of use of AI in terms of, you know, machine downtime, predictability, and all of that. I have been actually doing a lot of work in the sales and distribution and sales and marketing space, you know, with respect to AI, where there was a significant amount of data that comes from the connected vehicle. And there’s a lot of meaning that you can, you know, look into it, right. And then obviously the after sales part of it, where, you know, the vehicle is in the hands of the owners, and they are driving around, and there’s a significant data that’s coming from the vehicle itself, you know, all that put together. I think the role of AI is significant, you know, in the core functions with the manufacturers.

– There is a lot of data and that, you had mentioned also in your answer about historical data. So we have this legacy data that’s also available. We don’t just have the new data that’s coming once the users or for the owners are in the car and driving the car. We also have that. Can you tell me a little bit about how OEMs interacts with that legacy data? What are their opportunities? What can they do with that?

– Yeah, you know, that’s an interesting question again, because, you know, if you really look at, you know, AI and machine learning has a lot to do with data, right. And in fact, you know, the way I think this is, you know, traditional OEMs or manufacturers, I wouldn’t call them legacy OEMs, but you know, traditional OEMs, there are a lot of traditional OEMs who are big in this particular space. They’ve got a lot of historical data. You can call it legacy data or historical data. There’s a lot of meaning that can be derived out of data, right? So that is a huge advantage for the likes of both traditional OEMs. And then there are these new players, like, you know, Tesla and Vivian who are trying to kind of gather data from the consumers now. If you really look at it, you know, and I don’t know if I heard it or you don’t have started talking about it myself, Tesla is not a car company. I think you know Tesla is all about data. You know, they are learning like what Google did very, very early on. You know, you put the hands on the, you put the Tesla in the hands of a consumer and the driver is driving around and Tesla is collecting all that data. Right? So then a significant amount of data that gets generated, you know, both, you know, traditional OEMs as well as the new, new, new age OEMs like Tesla. And there’s a lot of meaning that could be derived out of that particular data.

– And this data, let’s say that the OEMs have at their hand, there’s, there’s a lot of potential with it, but it needs to be normalized. There’s a lot of opportunities, let’s say use cases, whether that’s, you know, mobility services or customer relations that can be derived or in our case with Upstream, you know, we normalize that data to be able to use it for cybersecurity reasons. How do you see that playing with the normalization of data, the use cases, or, you know, cybersecurity vendors like Upstream, how do you interact with that? How do you see the future or the growth of that from your perspective?

– Yeah, there is tons and tons of data that’s coming from the vehicles, right? As you know, you know, today there is a lot of the data that gets captured from the vehicle because the vehicle is equipped today to kind of classmate that particular data. Anyway, new vehicle that is getting produced or manufactured today is kind of, has an ability and a sensor to transmit that particular data. And if you really look at it, there’s a lot of data actually, you know, in, in some sense, I think, you know, you’re talking about, you know, very well about the normalization. If you really look at it, you know, there is, you know, you can capture every click and every element of, you know, or, or a trigger coming from the vehicle, right? You go to the car, you open the car, you close the car, you click your ignition. Every bit of that data is getting captured. But the fundamental question is, is that all required, right? In fact, does, that’s lots and lots of data. But coming to your question about the use cases, I think we are seeing a significant driver of use cases, even though I would say many of them are in a proof of concept stage, but after normalizing the data, there’s a lot of meaningful insights that we can derive, right? I will take a couple of examples that we have worked with OEMs, where for example, I can take a look at the data and see, if obviously the consumers have opted in for sharing that data. That’s very, very important from a privacy standpoint, I know if one of the OEMs, you know, owners of the vehicle is, there’s a thing, a competitive dealership, right? I can map that particular data, and know that he or she is driving that vehicle at what time.

– Seems like stalking.

– Yeah, you can say it’s stalking, but I don’t know if, if the meaning of that word is talking machines, you know, with the consumers opting in and the benefits that, you know, they are realizing. But, but we know everything, right? Where are they visiting? When are they visiting? Who are they visiting? How are they driving? What are the driving habits? All of that, is you know, getting captured. Now with that, actually there is a significant amount of, you know, to your point, risks also in work, right? The amount of data that’s coming, it’ll obviously you have very, very needs to be very secure. It has to be kind of very well, you know, warranted or whatever it is, right. But at the same time, often, we have also seen, you know privacy coming into picture in a big way. It’s not just the data about the vehicle as I talked about, it is now the data about the consumers. And you’re seen in the laws of like GDPR that came about in Europe, and then there was a CCP that’s, that’s in U.S. So I think manufacturers and OEMs today have to, spend a significant amount of time and energy and money, in terms of securing that data and making sure that it doesn’t get out or get attacked, right? Because this data, as I said, it’s not just about the vehicle. It will also hold the consumers, and there could be a significant damage that will happen.

– It’s interesting, that kind of double edged sword with the data and the fact that you have to both secure the data, but in our case as well, in Upstream’s case, we can use that data to actually secure the customers as well. So it’s very interesting to look from both perspectives at this engagement, with this data and within this connected vehicle space. Any last things that you have, you want to, you know, mention to our viewers, or you want to kind of give us the last statement here.

– Yeah, no, I think this was an interesting space Fay, and glad, you know, I’m talking to you. In fact, you know, I have been following, as I said, automotive for very, very long time. I’m, I’m very deep into it. And I think, you know, there’s a lot that is happening in this particular space. In fact, I’m happy to see actually that the national highway transportation safety association is in fact putting a significant focus on cybersecurity. They have a dedicated focus on this. And in fact, they have adopted, I believe the NIST framework and a multi-layered architecture that they are proposing working with, you know, not just the manufacturers and the Williams, but partners like us and the academics and all of that. So there’s a significant focus that’s happening. And we are also kind of very happy to partner with very, very niche players because equal system was very, very important with the likes of Upstream, you know, to deliver value to our consumers.

– Especially now with both in this framework, as well as the UNDC UWP 29 regulations, the upcoming, you know, 21434 standard from the ISOSE. There is a lot of changes that need to be made, and we’re really excited to kind of see that shift and understanding the value of data, and understanding the value of connected vehicle for OEMs and the likes. So I wanna thank you so much for joining me today, and glad to have recorded this with you.

– Thank you Fay, it was great talking to you.

– Pleasure.