Accordion

• Transcript

  Tyler Suiters  0:11 

  Hey everybody, Tyler Suiters with the Consumer Technology Association. We are the owner and producer of CES, the most influential tech event on the planet. We are here to get you CES ready. Big show's coming up January 8-11, 2019, in Las Vegas. And today we are taking on AI: artificial intelligence. It is one of the key ingredient technologies over the next decade. Things like biometrics, whether it's voice or facial recognition, predictive analytics, you can read that as big data and everything that goes into that, and self-driving vehicles, right, the future of transportation for us.

  Tyler Suiters  0:55 

  Well, CTA research says that across 12 industrial sectors, AI means big business. Companies adopting AI tech in those 12 industrial sectors all report higher profit margins. And this also applies to consumers as well. We're getting more familiar with AI through voice controlled smart speakers. And CTA research shows 64% revenue growth here in the U.S. in 2018 on the sales of smart speakers. And that's after only three years on the market. So today, three global innovators I bet you know well, who are getting more and more involved in the AI space. First IBM, who has put a name to artificial intelligence and that name is Watson. Intel will be discussing both the personal and the societal benefits that AI can deliver. And also a conversation with Yamaha. You probably know the company for musical instruments or motor vehicles. And we'll talk about the latter there but also some very sage advice for startups in the AI space. All of that is on this edition of CES Tech Talk.

  Tyler Suiters  2:07 

  Michelle Boockoff-Bajdek is chief marketing officer with IBM Watson AI. What an exciting place to be Michelle, thanks for joining us today.

  Michelle Boockoff-Bajdek  2:16 

  Oh, well, thank you for having me. It is an exciting place to be.

  Tyler Suiters  2:21 

  Part of that is the potential that AI holds, right, the things that it can do that it may do. Give us a quick level set, if you would, where is reality right now? And what is much more conjecture, right, the real AI. What does that mean?

  Michelle Boockoff-Bajdek  2:37 

  Yeah, you know that? That's a good question. And I think sometimes it's hard to separate the reality from the hype, right? So let's talk, let's level set on what AI is. So we're talking about artificial intelligence, which is what gives machines the power to learn, adapt to new inputs, and ultimately help us make better decisions. Now, machine learning, which is a subset of AI, analyzes data to make intelligent decisions based on what it's learned. And many AI systems, including our own at IBM, which is called Watson, you've probably heard of it, uses a sophisticated machine learning technique called Deep Learning. And deep learning enables AI systems like Watson to ingest vast quantities of unstructured data, something like 80% of the world's data today is still unsearchable. Things like photos and videos and audio files. Watson continuously learns on the job and constantly improves the quality and accuracy of the results. And so from there, you know, we as people can make better decisions.

  Tyler Suiters  3:38 

  So part of that hype or that that mystery that's out there is AI working with humans, right? And there is certainly a Hollywood-esque narrative that's been around for decades about the power of AI and what that means for us.

  Michelle Boockoff-Bajdek  3:55 

  Right.

  Tyler Suiters  3:55 

  Where do you see it in terms of AI as an assistant to technology, or a superlative technology in working with us as humans?

  Michelle Boockoff-Bajdek  4:05 

  Yeah, you know, it's a great question. Look, we do believe that AI will change industries and the way businesses operate. But we also believe this is about human and machine working together. The purpose of these technologies is to augment man not to replace it. So let me give you a few quick examples, because I think this will also show how Ai is enhancing the work we do. So in the case of Watson, currently, it's being used to assist customer service agents answer and respond to consumer inquiries. It's helping lawyers draft legal documents and filings, saving them time and money. And in both of these cases, it's helping these professionals work more effectively, and allows them to focus on higher value tasks. Now, here's the one you're probably going to like, Watson is even being used by people who play fantasy football, to help them make better decisions about their teams. Because Watson reads and comprehends and interprets millions of news articles and social content that might impact your players. And so you can actually improve your fantasy outcome.

  Tyler Suiters  5:12 

  Results driven AI in the present. Love it. Yeah. What about the business side? I mean, that's where IBM is so well known and has been such an innovative leader for decades and decades. The idea of AI as a black box, right? If businesses are building on using AI to make decisions, how do they in turn or parallel to that ensure that that level of trust is there too, right?

  Michelle Boockoff-Bajdek  5:41 

  Yeah, you know, that's so critical. Look, the only way that businesses are truly going to scale their AI projects is if they trust the outcomes that the systems are producing, right? And the way you achieve that trust is by ensuring fairness in your AI models, because when you have a model that is steeped in fairness, it's going to lead to better outcomes. So detecting and mitigating bias is one part of this. But there's also an element of transparency into the systems, right, you called it a black box. We need to demystify that black box and ensure that we've got full transparency into the decisions that these models are making, the recommendations that they're offering, while also ensuring that the underlying models and framework are incredibly secure.

  Michelle Boockoff-Bajdek  6:34  

  I want to go back to this notion of bias for a minute if you'll indulge me, because bias has many dimensions, right? So when people think of bias, we often think of societal attributes, right? Gender, age. But because AI systems comprise models that are trained on data sets, resulting biases and systems are likely not because of malicious intent, but rather the inherent bias in that data set. So let me give you an example, mortgage application data. So let's say that you were designing an AI system to help approve or deny applicants for large mortgages. Well you probably trained your underlying models on a set of historical data, let's say the last 50 years of historical mortgage data, it's likely that more men than women were probably approved for mortgages during that period, because the historical data skews towards men.

  Tyler Suiters  7:29 

  Well, if you're going back to the 60s and 70s, absolutely.
   

  Michelle Boockoff-Bajdek  7:31 

  Exactly. And so what's going to happen is if your data is skewed, it is likely then that you are going to make different decisions. So maybe you are starting to approve more men than women for these mortgages, because again, the data suggests that. But that is an inherent and unconscious bias in the system. And we have to be able to step back and mitigate.

  Tyler Suiters  7:53 

  So for people who aren't immediately in the AI space, Michelle, it seems like the Television ads about Watson are the most immediate touch point for many people to learn about what Watson is and the opportunities there. Examples, Watson, as you called out earlier, is used at a call center, something that I think you could understand at a consumer level. But then Watson's usage goes so far down the spectrum. We're talking about, you know, like proactive maintenance of airplane turbines, which is far beyond consumer usage. But those are just two examples. Where else is Watson being used right now that we can understand and relate to?

  Michelle Boockoff-Bajdek  8:33 

  Well, we could we could be here for a long time.

  Tyler Suiters  8:36 

  Fair enough. Fair enough. And I'm going to 30-second commercial by 30-second commercial, so.

  Michelle Boockoff-Bajdek  8:43 

  You know, Watson is all about helping people accomplish more in their daily work. And we have so many Watson products that are pre trained for specific industries and professions, just like you mentioned, right? So whether we're talking about aviation, or we're talking about legal profession in the example I gave before. But I want to come back to call centers for a moment, because this is one of those areas where I don't think that people realize the value that Watson is bringing to some of the experiences that people have.

  Michelle Boockoff-Bajdek  9:15 

  So customer service agents in call centers, they need information right away when they are having- when they're engaged in the conversation with a customer, and they want to be able to provide the best answer to customers, but also speed response times. And so Watson is helping them. So example, Bradesco is one of the largest banks in Brazil, and has used Watson to support 10 million customer calls, with 85% of those customers evaluating the interactions they have as good or excellent. And then another client Eubank sped up processes for its customer agents. And this is going to be somewhat mind boggling, but they improved response times for more than 400,000 customers, and cut down search tons by 33%.

  Tyler Suiters  9:59 

  Yeah, I want to do a full stop. Pause right there, Michelle? 33% reduction. Really impressive in today. Yeah.

  Michelle Boockoff-Bajdek  10:10 

  Well, and if you think about it from a customer experience perspective, you as a customer are going to feel so much better about that engagement, and in the brand you're working with, right?

  Tyler Suiters  10:18 

  Mm hmm.

  Michelle Boockoff-Bajdek  10:20 

  But you know, we're not stopping there. Watson is also embedded into the technology platforms that we see a lot of professionals using today, whether it's Box or Salesforce or WorkDay, and we're bringing AI to those systems and the workflows that people are already using so it's going to feel natural when they can start to engage with artificial intelligence, and it's going to give them more insight into what's happening with their customers. In the example of like a Salesforce.

  Michelle Boockoff-Bajdek  10:49 

  Another area we see AI emerging is in advertising. And this is brand new in a first of its kind for the airline industry. Lufthansa if you can watch an ad- Now this is an any high-powered advertising format that lets consumers engage in totally new ways, and have a two way dialogue in natural language with the brand.
  So in Lufthansa's case, potential travelers via the ad can ask the grand general travel questions. What's it like to fly with Lufthansa? How do I check in for my flight? And based on the consumer's interaction with the ad, Watson will serve up travel tips, vacation ideas, but also include a link to the flight reservation page. These are really some unique ways in which Watson is helping brands do so much more.

  Tyler Suiters  11:35 

  Very cool. Michelle, what about the next step, scaling all these implementations? What do enterprises need if they're going to grow their AI projects and really leverage this new AI technology? Not just have it on board and be able to check the box to say, yes, we're AI enabled, but really get the most out of it?

  Michelle Boockoff-Bajdek  11:54 

  Yeah, you know, it's such a great question. We see with many of our client companies that AI is moving beyond the pilots and the simple implementations to much larger scale deployments but it's still challenging. So to support this, we're excited to announce that we launched this offering called IBM AI Open Scale, which is designed to help companies accelerate the adoption of AI in their enterprise in a few crucial ways.
  So first, and there are three specific ways, first, AI Open Scale will help companies provide explanations into how AI decisions are being made. So we talked a little bit before about fairness. Well, AI Open Scale will automatically detect and mitigate bias to produce the fair and trusted outcomes that we talked about. If you can't trust the recommendations the system is making, you won't rely on it. And the same is actually true when we're talking about AI in your organization. So that's number one. So it provides that explanation.

  Michelle Boockoff-Bajdek  12:58 

  Second, it applies automation to your AI. So AI Open Scale includes something called neural network synthesis, or what we call newnets, which allows data scientists to upload their data, and then newnets designs and builds a model for them from scratch. Now think about that. This is AI building AI. So that's number two. But number three, the key to AI Open Scale is in the name. "Open." You need to be able to work across a range of tools, frameworks, and environments, and support your models, wherever they're deployed, whether it's in the cloud or your own data center. And that's why we created AI Open Scale. It's open by design, meaning that it integrates with common AI tools, frameworks and environments, across clouds, public, private, or on prem.

  Tyler Suiters  13:53 

  So what about the next step? And I don't mean just the year ahead, Michelle, but but the years plural. IBM CEO Ginni Rometty is delivering the opening keynote at CES 2019. I'm sure she'll lay out her vision, the company's vision, for where AI belongs. But what about Watson? How about a little sneak peek at what's up next?

  Michelle Boockoff-Bajdek  14:15 

  Yeah, well, I'm certainly not going to steal Ginni's thunder.

  Tyler Suiters  14:17 

  Not going to ask you to, no, no.

  Michelle Boockoff-Bajdek  14:21 

  I really do want to talk about one of the hidden gems, one of the greatest assets in building the future vision of AI or Watson, and that's IBM Research. They've been a huge driver of our innovation in AI. And one of the key reasons that IBM has been the top U.S. company for patent Awards The past 25 consecutive years.
  Our inventors receive more than 9,000 patents in 2017 alone. That is remarkable. But one of the most exciting new breakthroughs in AI is a system called Project Debater. I don't know if you saw but back in June, it engaged in the first ever live public debate with humans.

  Tyler Suiters  14:22 

  Right, Right.

  Michelle Boockoff-Bajdek  14:53 

  So this system, it goes beyond search based discovery technologies that we're used to, to working with humans to discover reason and, here's the key: present new points of view. So now let's think about how do you use this? Well, businesses can use this technology to understand multiple perspectives on a decision allowing for more nuanced reasoning, and prompting users to really challenge some deep set beliefs. So for example, financial advisors could use procon analysis to surface facts that would help them find the right product for a client or a lawyer preparing for a trial could use summarization to research the right legal precedents to take to court, as well as test the strengths and weaknesses of a case using a mock legal debate. And this is only the beginning for what we believe is truly new, exciting technology that's coming only from IBM.

  Tyler Suiters  15:56 

  Okay, so Michelle, Debater, I understand we're going to see that as part your presence at CES 2019. That's an exciting way to kick off.

  Michelle Boockoff-Bajdek  16:06 

  Absolutely. Well, you can certainly come by the Las Vegas Convention Center. We're going to have an incredibly large presence there, you'll see Debater and you'll see quite a few more exciting things that we have planned. But I would say to everyone don't miss Ginni Rometty's keynote of course. That I think is going to be the highlight of the event.

  Tyler Suiters  16:24 

  Alright, Michelle Boockoff-Bajdek is Chief Marketing Officer with IBM Watson AI, exciting times at IBM and I'm sure much more to come. Michelle, great to have you with us.

  Michelle Boockoff-Bajdek  16:35 

  Oh, thank you so much. It was a pleasure.

  Tyler Suiters  16:40 

  Julie Choi is with Intel where she is Global Head of AI. Julie, great to have you with us today. Thank you.

  Julie Choi  16:47 

  Thank you so much, Tyler, great to be here.

  Tyler Suiters  16:49 

  So let's talk a bit about your role. Global head of AI, I get, global head of marketing, I understand. But you put those two together and it's an indication of the emphasis, the premium that Intel is putting on AI as a business strategy right now.

  Julie Choi  17:06 

  Absolutely. It's such a privilege to be working on AI marketing for Intel. As many of us know, AI is becoming a huge part of our lives, in every experience. And through the development of our software and our hardware and processor technology, Intel is really finding ourselves at the heart of powering so many of these new types of AI experiences. And it's my job and pleasure every day to look for exciting examples that take AI from theory to reality, through the stories that our partners and our collaborators are building every single day.

  Tyler Suiters  17:48 

  So what are some of the stories you're seeing that are really catching your attention and amazing you right now?

  Julie Choi  17:55 

  Yeah, there's just so much going on in the world of AI and so much innovation, both in the science and in the application of the science to many different industries. I'm really excited to see AI touching and transforming industries like health care and medicine. But it also is relevant to just every day, you know, experiences like online retail and the search that's powering that, or the recommendations that we see in our social networks, or even the graphics that we're seeing, the  amazing creatures that we see in movies. So it's just thrilling to see how AI is becoming a part of every aspect of our lives in this way.

  Tyler Suiters  18:43 

  So you would certainly agree that AI is already impacting our lives today in large ways and small, we cited some of those. How do you define that? What is an impact through AI on the way we're living our lives now or little elements of what we do every day?

  Julie Choi  19:02 

  Sure. So why don't we talk about some examples, right? So one element of my life every day that I'm sure people might be able to relate to is search, right, searching for information on the internet. And one of the partners that we've been working with over the past few years is Microsoft, and specifically on powering the way that their search engine Bing is delivering world class search results to the end users of Microsoft Bing. And the way that that's happening is through the processing of the information on Intel FPGA technology. And specifically, it's in collaboration with Microsoft Bing and a platform that Microsoft has built for AI called Project Brainwaves. And Project Brainwave runs on Intel FPGA to deliver, to basically enable Bing to quickly process millions of articles across the web and deliver relevant answers to the users of Bing.

  Tyler Suiters  20:16 

  What about the bigger picture, some of the larger societal challenges we're facing? I think of smart cities, of resilience. And that trickles down a bit into drones or self-driving vehicles. But what about those bigger picture ways that AI is not just engaging or improving their lives but really solving the challenges we face today?

  Julie Choi  20:41 

  Absolutely, I mean, when we talk about the big picture challenges, I think what's really exciting to me is the innovation that I'm seeing in the delivery of healthcare, you know, to end users like patients in the ER, or in the ICU. And Intel, we've been able to partner with companies like Accuhealth, a Chilean based startup that's developed a patient centric healthcare model that really enables faster analysis of disease patterns so that patients can receive care and reduce the amount of visits to the ER.

  Julie Choi  21:26 

  And so that's just one healthcare example, from Accuhealth. But we see other examples in health care that are truly impactful, such as with Montefiore, which is a nationally ranked health system, largest in the Bronx. And they've created an AI platform called PALM, that stands for patient centered analytics and learning machine. And they're able to take the data through the PALM system to reduce patients time in the ICU. And so in these types of ways, we're seeing AI solving really large problems in healthcare and delivering diagnoses and helping patients faster than ever.

  Tyler Suiters  22:11 

  So that's an interesting point, Julie. Is that where AI is going to be incorporated in I say, virtually every industrial sector, but maybe it is every single one. And that is the issue of data analysis, right, the gathering, but more importantly, the calculations that go around these almost infinite points of data that we're beginning to capture, but can't begin to analyze in an effective manner without the help of AI.

  Julie Choi  22:39 

  Yeah, absolutely. I mean, AI, if you want to just think about it in the simplest way, artificial intelligence is about taking data. So large amounts, different varieties of data, different velocities of data, I mean, we're all creating data more than ever, through our mobile phones, or iPads, our watches, our sensors that we're wearing, when we go on our runs. I mean, data is coming from everything in our lives, even our vehicles. And what artificial intelligence is, is it's the method, it's the method for taking all of that data. And, you know, through the help and expertise of humans, data scientists, we call them, or data engineers, we actually turn that data into really useful predictive models. So we shape the data into a structure, into a model that can basically then inform computer programs all this data indicates this pattern.  And then we can teach systems to basically anticipate situations faster than humans could, because it's just such a large amount of data. And we now need computers to help us actually understand what that data is implying about disease, about maybe what I want to eat, what I want to wear, or who is popping up on my newsfeed on an application like Facebook?

  Tyler Suiters  24:24 

  So I'm going to ask for a bit of a preview, and maybe some inside information, Julie, if you don't mind, but Intel's keynote at CES 2018 was spectacular, the talk of the town after that presentation. What do you all have in store for CES 2019? Because gotta believe they will be the focus and have to believe it's going to be big.

  Julie Choi  24:47 

  Yes, absolutely, we are so excited Intel is so excited to be a major part of CES 2019. And I mean, we have a great show planned for people that come and who want to learn more about what Intel is doing in AI, but also other areas like 5g, and, just the data center and IoT. There's really not a part of the world that Intel isn't involved, because, of course, our processors are really powering the ability to make sense of the data that's flowing. And at CES we'll basically be collecting the best of, the latest that we're seeing with our customers and the partners that we work with, in terms of the applications that they're building, and how we're working together to process their data, to create experiences that are just really transforming our world.

  Tyler Suiters  25:46 

  Alright, so this admittedly is a bit of an unfair question, maybe totally unfair, Julie, but can you give us a peek behind the curtain for what will be on the show floor?

  Julie Choi  25:58 

  I think, preview in terms of a peek behind the curtain, you'll definitely be seeing amazing and new types of silicone products from Intel. And then applications that they'll be powering probably in cars. So we'll definitely have updates on autonomous driving and how we're powering that intelligence there. And important industries, again, that are impacting everyday problems, like how we're partnering with healthcare providers and doctors to deliver advanced analytics for patients, or maybe on the retail side as well.

  Tyler Suiters  26:45 

  All right, that is certainly enough to get us amped up about what you all have planned at CES 2019. Julie Choi, Global Head of Marketing with Intel. Such a pleasure, Julie, thank you so much. And we'll see you in Las Vegas.

  Julie Choi  27:00 

  Thanks so much, Tyler.

  Tyler Suiters  27:03 

  Joining us now from Yamaha Motor ventures where he is a partner is Amish Parashar. Great to have you with us from the west coast today.

  Amish Parashar  27:11 

  Thank you for having me.

  Tyler Suiters  27:12 

  No, very kind of you. So AI and motor vehicles, we hear that inexorably connected with the future of self-driving vehicles, right?

  Amish Parashar  27:22 

  Yeah, absolutely. You know, it's kind of a fundamental thing. It used to be that vehicles needed to have pistons and crank shafts. And now we're moving into an era where without AI, maybe a vehicle might not be possible to produce.

  Tyler Suiters  27:37 

  That is a bit of a transition, right? No steering wheel and then absence of a lot of other things. Maybe in the cabin, we'll see. Where else is Yamaha taking this from the AI standpoint? You're so diversified in terms of getting beyond just passenger vehicles. Where else are the applications? What are the possibilities there?

  Amish Parashar  27:57 

  Yeah, so we view it as a kind of a manifold problem here and a manifold opportunity ahead of us. And as you pointed out, Yamaha makes vehicles that that cover land, sea and air, especially here in the U.S., a number of our vehicles are used for recreational purposes. We hope that people buy or use our products, get into the outdoors, and then have some adventures. And maybe once in a while associate our brand with those passions and those adventures. You know, for us, we can take something like a motorcycle as an example.

  Amish Parashar  28:30 

  We have experimented with self-driving motorcycles, we've shown some at CES. The act of a rider on a motorcycle is this connection between a person and a machine that's kind of hard to describe in words, right? We have an emotional reaction to machines like that, as human beings. Getting rid of that riding experience is not our intention, right? But enhancing It is something we hope AI can help us do. That's just one example.

  Amish Parashar  29:00 

  We feel like self-driving vehicles of all kinds are coming at us pretty quickly here. The major automakers are trying to create products and vehicles that will allow us to go from Los Angeles to New York, fully autonomously, all weather conditions, all road conditions, all times of year. We're taking a slightly different stance and saying that local mobility is changing very quickly. And that we think lower speed local mobility will be the first to adapt self-driving technologies. And so we have a number of efforts underway, including some we'll show at CES 2019, that are focused on local navigation and local autonomy, with, of course, AI built in to enable some of those.

  Tyler Suiters  29:48 

  Well, let's talk about those manifestations. I mean, what will we see when you say this is some of the first AI enabled transportation technology that we as consumers will see? Paint a picture for me about where Yamaha sees us going in the near term.

  Amish Parashar  30:03 

  So maybe I could start by doing the opposite, right? One of the hardest things we can possibly do in terms of a self-driving vehicle is trying to account for all possibilities, right? That means a country road or farm road, it means a 10-lane highway, it means motorcycles and bicycles sharing the lanes with cars, it means trucks, it means bad weather and good weather. Trying to account for all of those possibilities that we do as human drivers is an extremely high bar for an autonomous driving system. So we've looked at that problem and said, You know that's coming, it's probably a few years away before that's widely implemented. Of course, that means we need to start today on activities like that. And many of those activities have indeed been underway for some time.

  Amish Parashar  30:55 

  But we look at that, and we say, you know, they're shorter term opportunities for self-driving that we'll see right around us, and we're starting to see already in commercial ways. For example, getting from my office to the local train station, about a mile on local roads, the speed limit scientists something like 25 miles an hour, it's about a mile of distance. It's relatively straightforward to map out that route, to map out alternative routes, and to start driving autonomously in that more predictable environment than trying to account for all possibilities. So if I say it's a shorter distance, and it's a lower average speed, we'll see AI, you know, in the form of self-driving vehicles implemented much quicker. And so that's kind of one direction we're headed.

  Tyler Suiters  31:43 

  Well, so part of your scope at Yamaha Motor ventures is on the VC side, right? On looking for the most innovative startups and making business decisions around that and the envisioning that goes along with that. What has you excited right now in the AI sector?

  Amish Parashar  32:01 

  Oh, absolutely. Yeah, that is part of our responsibility here is the corporate venturing team for Yamaha. You know, what has me excited is really precise applications and uses of AI. And we're seeing it in several ways. We have some effort underway on our team related to the future of agriculture. And so AI robotics, and agricultural harvesting, for example, is a really interesting combination. Again, it's a confined area on a farm or orchard. And AI allows farmers to make better decisions, or robots to help with very labor-intensive tasks such as harvesting or pruning. So we're really excited about that, those kind of possibilities. We're excited about other very precise applications. So things like goods or people delivery in small lower speed areas, for example, in urban center, a downtown area, or a college campus. So we like the real personal application of AI. I'm getting less and less patient with general purpose AIs, or AIs that are many years out in the making. I feel like we have real applications today and that we should be backing those in an aggressive way.

  Tyler Suiters  33:15 

  So from that perspective, Amish, what is your general guidance or strategic advice to smaller companies, to startups that are in that space and looking for major international partners, groups like yours, that are in AI and want to know what direction to take? It sounds like you're saying more specific applications are far more attractive. But is that too much of a generalization?

  Amish Parashar  33:40 

  No, I don't think it is, I think it's very attractive. We do invest in in companies that are kind of pre revenue, right? And so I won't benchmark it on revenue generation. But if there's a clear path to revenue from a specific set of customers, that's much more interesting than very advanced technology that may never see the light of commercial day. So for example, if somebody says, look, we can use these self-driving vehicles to be able to deliver ecommerce packages from a distribution center to the community that's a mile down the road and we already have a few potential partners that we're in talks with. To me, that's much more compelling than saying we have a machine vision system that can do a little bit better than everybody else on the highway. Right? The technology enables the business in most cases for us, and we're excited about those possibilities, more than the pure technical achievement.

  Tyler Suiters  34:37 

  So where are you looking for those innovative startups at CES? Do you just beline to Eureka Park? What's your approach there?

  Amish Parashar  34:46 

  Well, we have a lot of fun in Eureka Park, you know, inevitably we see a lot of old friends and a lot of partners. You know, we're lucky to be right in the mix of things here in Silicon Valley. And then our travels take us pretty far and wide. We have we have a team in Australia that's working hard down there. Of course, we have a team in Japan that covers Japan and a good chunk of Asia. I just got back from Europe. We're scouting globally, we are a global company, our core products are sold in something like 200 countries. And we view our responsibility as global.

  Amish Parashar  35:20 

  And so yeah, you know, Eureka Park has a lot of old friends and a lot of good contacts hanging out there. And we spend time there. A number of our investments are done with Co-investors. And those co-investors are well known Silicon Valley VCs, they're other corporate venturing groups, and generally friends. And so we spend a lot of time with those folks in their booths and in meetings that they're hosting or that we're hosting, that we pull them into. One of the the immense powers of CES is that your network ends up being in the same place at the same time, which kind of, to me only happens about once a year.

  Tyler Suiters  35:58 

  I would say precisely once a year, usually the second week of January, but yeah.

  Amish Parashar  36:02 

  That's right.

  Tyler Suiters  36:05 

  So while we're talking about CES and specifically 2019, Amish, what does Yamaha Motor Ventures have planned? What is your game plan for the show?

  Amish Parashar  36:15 

  Yes, so a few things. You know, it's quite an exciting year for us. You know, Yamaha's thinking about not only this year right now, but also thinking about what the future might hold, internally looking as well as externally.
  And so you know, the first thing you'll see is the Yamaha booth is located in the AI and robotics section again, this year. We're, of course, a major vehicle manufacturer. But we made kind of a strategic decision to highlight some of our robotics efforts that have been going on for decades, to highlight some of our interest in AI, to highlight some of our land, sea, and air kind of efforts.

  Amish Parashar  36:55 

  And so, come look for us in the robotics and AI section. I think it'll be kind of hard to miss our booth, you'll be able to see that tuning fork logo, I think for miles away. Stop by and say hi, we have a great team that's manning the booth there. We have a special couple of surprises in store that I'm not sure I can tell you too much about right now. But we hope to show that, as I said earlier, that self-driving vehicles are here and are useful right now today. And so I think if you go outside the exhibit halls, you'll see some evidence of Yamaha vehicles being used in really practical, near term ways. And I hope that inspires some thoughts for entrepreneurs, startup companies, even large companies to see the possibilities.

  Amish Parashar  37:41 

  And yeah, so at least those two things. And then there may be a few of us Yamaha folks on stage in various roles, trying to kick off some conversations about the direction we think the world is headed or should be headed.

  Tyler Suiters  37:55 

   All right, speaking of attention grabbing CES 2018. So the show this past year. I'm going to say one word, and I'm going to let you dive into it and give us an update. And that one word is Motobot.

  Amish Parashar  38:12 

  One of my favorite projects.

  Tyler Suiters  38:14 

  Yeah, Amish, go.

  Amish Parashar  38:16 

  I hope you have a couple of hours.

  Amish Parashar  38:18 

  Motobot is great, and I would encourage the listeners to actually go to YouTube, and type in Motobot. And if a picture's worth 1,000 words, maybe the video is worth 10,000 words.

  Tyler Suiters  38:31 

  Yeah, and I bet a lot of people are going to recognize because it got such great traction. And if you saw it at CES 2018, you did not forget.

  Amish Parashar  38:40 

  Great. Thank you for saying that. It's one of my favorite projects here. So Motobot, for those that haven't seen it at CES or online, is a humanoid robot that rides an unmodified Yamaha motorcycle. We got a Yamaha R1m one of our flagship models, and put this humanoid robot on with the stated mission of doing two things: one, exceeding 200 kilometers an hour in a race track. And number two, beating the best human rider we could find.
  The best human rider we could find is actually a nine-time world champion Valentino Rossi, who rides a Yamaha for in Moto GP. He's one of the world's best riders, easily in the top few. And our goal was to have our robotic rider beat him. We fell short of that goal. Turns out he's extremely good. But we did pass 200 kilometers an hour and a racetrack.

  Amish Parashar  39:35 

  And you know, just from a gut sense, you know, Motobot, the robotic humanoid rider is better than I am on a racetrack and better than most humans probably are. It's a wonderful demonstration of what robotics, artificial intelligence, machine learning, some simulation, and a bold ambitious project can do to kind of advance the state of the art. We're happy to see that many of our competitors in the motorcycle space are tracking this project and are working on similar efforts now that that project started about four years ago in rapid fire.
  You know, the first test runs I was ever at were on a disused runway going a few miles an hour. And the big question at that time was would it stand up by itself? Or would it be able to do a few miles an hour on its own without a human being? And then you know, the team was able to check off a whole bunch of boxes in rapid succession to the culmination is the videos you'll see online of the race between Valentino Rossi and Motobot. The results speak for themselves, you know, we wish we could have exceeded his laptime. But if you watch the video, I think we're within shot of doing it.

  Tyler Suiters  40:47 

  You know, we often talk about the lightning fast pace of innovation. Seldom is it actually as fast as lightning as you're describing to me. Well, a lot to look forward to at CES 2019 from Yamaha Motor Ventures, where Amish Parashar is a partner. Hey, Amish great convo today and thanks for getting us so excited about AI at CES.

  Amish Parashar  41:11 

  Oh, and I'm looking forward to it. You know the one of the joys of being there is to be able to walk around the area, see what everybody else is up to and what's getting them excited. I always view it as such a tremendous learning opportunity. And thank you for all of you. All of your efforts and everything CTA is doing to make sure CES 2019 is one of the best ever. I'm really looking forward to being there and hope to meet many of the listeners there as well. Thank you so much.

  Tyler Suiters  41:34 

  We very much want you to be CES ready, so a good first step: subscribe to this podcast so you won't miss any episodes. The show itself, CES 2019 is coming up January 8th to the 11th in Las Vegas. The information you need to get ready is at CES dot tech. As always, none of this is possible without our resident superstars: engineer John Lindsey and producer Tina Anthony, you are both the very best.
  Thank you. For this edition, I'm Tyler Suiters. Let's talk tech again soon.