Robots – Enhancing Our Security – Part 2

Some Questions I Ask:

• Does being the son of immigrants influence you in any way? (05:51)
• How did you go from working at Ford to building Knightscope? (09:00)
• How will the K7 robot be different from the robots you already have? (21:04)
• Do you think those police departments will ever hire your robots? (23:36)

What You’ll Learn in this Episode:

• The events and circumstances that inspired the creation of Knightscope (01:39)
• The different types of sensors that Knightscope’s robots have (11:19)
• What the company does to deter malicious damage to their robots (13:37)
• The different flavors of data that their robot generates (15:52)

Connect with William Santana Li: 

• LinkedIn
• Knightscope
• Twitter

Connect with Ed Bernardon:

• LinkedIn
• Future Car: Driving a Lifestyle Revolution
• Motorsports is speeding the way to safer urban mobility
• Siemens Digital Industries Software

Ed Bernardon: The increasing demand for our police and security forces is real; they are overworked and under-resourced. But are robocops the answer? Can robots actually be used to improve our security? Robotics and AI tech is advancing at a rapid pace and now autonomous robots can be made to survey perimeters, collect pertinent data, and facilitate security operations. So how do we use robots to deploy a viable security solution?

Ed Bernardon: Welcome to Part 2 with Knightscope’s Chairman and CEO, William Santana Li. Knightscope is an advanced security technology company that wants to make America the safest country in the world with fully autonomous security robots built to deter, detect, and report. Bill and I talk about his career and his vision for how robocops fit into our future. We discuss what it was like growing up as the son of immigrants, his career at Ford, and how it led him to start Knightscope. He also provides a glimpse into the future unveiling of Knightscope’s next robot, the K7. Join me, Ed Bernardon, on the continuation of this episode of The Future Car as we take a look into how robots may just be the next step to improve our security.

Ed Bernardon: Now, your founding partner was a police officer. You were actually inspired, to some extent, by what happened at Sandy Hook back in 2012. Can you tell us a little bit about how you met your partner, what you were seeing that was happening out there really got you going on Knightscope?

William Santana Li: Going back a little bit further than that. I was born in New York City, someone hit my town on 911, I’m still profoundly ticked off about it. Back then I decided to dedicate the rest of my life to better securing our country. And actually, Stacy Dean Stevens had reached out to me roughly in the same timeframe and said, “Hey, I’m a police officer out here. We need some help. Looks like you’re trying to build some interesting stuff.” So, we’ve been at it for quite some time. Him being a practitioner and being out in the field, in the crossfire, and not having the tools or, said differently, he always talks about the toolbelt and massive amount of stuff that he has and “I can’t fit any more stuff and I still don’t have what I need.” So, I think that led to a lot of conversations. And then after Sandy Hook, it was like, “Okay, enough. How do we deter this stuff from occurring in the first place? How do we get more boots on the ground (for lack of a better way of saying it)? And how do we build some tools to stop the chaos?” The thing that ended up being extremely difficult is I think you cover autonomous self-driving technology is rather difficult, robotics is rather difficult, AI is rather difficult, and electric vehicles are rather difficult. For us, it’s all four into one. And oh, by the way, it’s not a science project, it’s not an R&D project; you need to be operating 24/7 in the real world with the client down your throat. That’s a whole other level of commitment and engineering required. But it’s been a long trek to get to this point. We’re nine years into it. I’m so excited that that first chapter is done and now we’re working on the second chapter, which is growing the company and expanding our footprint.

Ed Bernardon: You actually raised another good point, like you said, it’s more eyes and ears. But a lot of times, especially in these situations where you may have a shooter or something like that somewhere — you don’t know where they are — putting a robot in harm’s way to gather information is much, much better than risking the life of an officer to try and gather that same information.

William Santana Li: My CFO doesn’t like me saying this, but I’d just as soon have one of our robots get hit in the line of duty; the asset is replaceable, the human is not. We want to make sure that that capability over time is built out, but you need to take a layered approach to it. Everyone thinks you just need that one thing and it’s going to solve how this works. That’s not how society works. You need to put more boots on the ground, you need to keep the humans that you have compensated appropriately, you need to give them more tools, and you need to have a layered approach. One of the ones that drive me crazy is the airports; a lot of folks have flown. In the TSA, our friends over there spend an inordinate amount of money securing and hardening that checkpoint to get into the airport. Has anyone spent any time thinking about “Hey, did you secure the curb properly? Did you secure the parking structure properly?” How about let’s work on not getting the individual in the building in the first place? And that kind of thing is where I get really frustrated. In a lot of cases, it’s no one’s fault. There are too many people at an airport that are involved in the security footprint: there’s the city, the county, the federal authorities, the guarding company, the parking people. It doesn’t look like an entire system, it’s just like, “Well, I’m going to secure my portion that I have a budget for, that I got approved for. I don’t know about that parking structure because I’m not responsible for that.” And that lack of systemic thinking is a huge problem in our country.

Ed Bernardon: Again, this is very similar to autonomous cars. We had Mark Rosekind from Zoox on. He’s their Chief Safety Officer. And their focus is on preventative safety rather than reactive. And anything you can do — add eyes and ears or coordinate in a better way — to keep something that you don’t want to happen from happening is always a lot better than having to use a machine or whatever, like an autonomous car, to actually maneuver because an accident is occurring; you want to just avoid it in the first place. This is the same thing here. A little bit about your background. You were the son of immigrants; your mother’s from Colombia, your father’s from China. Did that influence you in any way as you were growing up?

William Santana Li: Ed, it depends on the meeting. I’m either very wise or very passionate, depending on which subject we’re on.

Ed Bernardon: My parents were both from Italy so I tend towards the passionate one. Did they impact you in any way? Do you think being the son of immigrants changed you in any way?

William Santana Li: I think that that always plays a part. But I had a little bit of an odd upbringing, it was not what you might think. So, my parents were the opposite of– I would get yelled at for studying too much. I would get yelled up for staying up till two in the morning. I would get yelled at for “Why don’t you come to watch the movie? Can you put the book down?” It was odd. I mean, my parents were absolutely awesome. They’re both blue-collar; my mom was a nurse’s aide and my dad was an inventory manager. I don’t know how they made ends meet. I made more as a summer intern than both of them combined, but they made it work. And I think that determination certainly made a difference. But also this country gave them the opportunity to even have a much better life than where they were. For that, I’m forever grateful. So, me dedicating the rest of my life to better secure this country couldn’t probably resonate more than that. But I don’t know where the drive to work crazy hours and keep at it, but once I set my mind on doing something, I will do whatever it takes even if I have to force it to make it happen. And I think, now, I’m profoundly lucky to be surrounded by a management team that’s willing to go to the mat as well. So, it’s usually not a good idea to bet against the Knightscope team.

Ed Bernardon: My father was just a cement finisher and my mother was a seamstress, and that determination and that drive, seeing how hard they were working just so they could take care of us, I think is what pushes you; it pushes you to stay up till two in the morning sometimes to get something done. And speaking of that, when you tried to get your first job at Ford, you sent out 60 resumes, I believe, and only two came back and both of them happened to be at Ford. That’s what I call lucky-lucky there. If you’re only going to have to come back, might as well be your top one. Talk about disappointment.

William Santana Li: Every automaker out there to try to get a summer internship, no one would have it. And then the number one choice that I really wanted, I get two offers. So, I got lucky.

Ed Bernardon: So, how do you go from working at Ford? There are a couple of jobs in between. You started a company to make a purpose-built police car, that’s where you met your co-founder of Knightscope, and then you went on to Knightscope. Just a little bit of that transition in there. Because probably Carbon Motors — that was the company where you had the purpose-built car — taught you a lot about what police officers need. 

William Santana Li: Yeah, I think Stacy and I had built 2500 relationships with different law enforcement agencies across the country, so we got rather fluent on the problem. I think what gave me the entrepreneurial bug was my last gig at Ford Motor Company. I was given kind of a special assignment to go build a subsidiary for Ford. So, ill-advised, I was an intrapreneur. I got the board to release a quarter billion dollars to me to do a roll-up in the used parts industry. I bought 22 companies in 11 months, had about 600 employees, doing about $150 million in sales. And that gave me that entrepreneurial bug of being able to go build a company– At the time, I had my own board, my own HR, my own treasury. So, then less excited to go back into the mothership to go be the vehicle line director for whatever XYZ vehicle when I’m sitting here running a company I just built. So, that kind of gave me the entrepreneurial bug. And I’ll tell you, Ford was an awesome training ground. If you go on my LinkedIn profile, you’ll look “unemployable” during my period there because every few months, I was getting promoted or double promoted to go fix some other problem. I think I had like 12 jobs in 10 years. 

Ed Bernardon: You were like a vehicle body engineer in there for a long time and then finally at the end.

William Santana Li: I was an electrical engineer, I was a systems engineer, I was a vehicle engineer, I did body engineering, manufacturing, rationalization, market research, M&A. I got to touch, feel, and see on four different continents, the inner workings of a massive company. A lot of good things there and a lot of not-so-great things. A couple of things I miss, a lot of things I don’t miss. But it was, again, an awesome training ground to understand the inner workings of a major automaker.

Ed Bernardon: You’ve touched a little bit on some of the sensors that your robot has, and they can work in low to no light conditions, they have facial recognition, thermal sensors, cameras. Tell us a little bit about the different types of sensors that you have, and how you fuse that data together for the robots to do what they need to do. 

William Santana Li: Probably need to split that question into two. One is the navigation side of things and the other is more on the security side of things. So, on the navigation side of things, very similar and analogous to a self-driving car, use a bunch of sensors to create a dynamic, real-time, 3D mapping of the environment.

Ed Bernardon: What sensors do you actually have? LIDAR, radar, all the same ones you see in an autonomous car or a subset of that?

William Santana Li: LIDAR, sonar, accelerometers, and a bunch of crazy software. We don’t ingest blueprints or anything like that; you’re dynamically mapping an environment. For us, it’s a little bit easier than an autonomous vehicle because we’re patrolling the same area over and over again; we’re not having to be at a random location, random time, random everything, which to me is kind of a difficult engineering exercise. And then on the security side of things is what you spoke of: using a good amount of machine learning to be able to detect a person do the facial recognition, run a thermal scan to see if you’re either trying to find something or trying to alert someone on a safety hazard. And over time, what we hope, and I’ve often said this for way too many years, we want the machines to be able to see, feel, hear, and smell and do 100 times more than a human could ever possibly do. And as sensor prices come down, as compute capability comes down, and the more we can do at the edge versus doing stuff in the cloud, those capabilities become much more viable. And then there’s definitely, as I alluded to, some sensor fusion. So, some of the stuff that you would be interested in from a navigation standpoint would obviously be interesting on the security side of things. I’ll give you a dumb example: “Hey, people are usually vertical, I saw a person that’s horizontal, let’s navigate around that person.” The security team would want to know that there’s someone horizontal. So, it’s stuff like that, that’s going to overlap over time. 

Ed Bernardon: We were talking earlier about the interaction with people. So, do you ever see any issues with “Oh, there’s a bunch of Knightscope robots around here. You run up to that one and cover the sensors with shaving cream or put a duffel bag over the robot? Or just knock it over maybe.”

William Santana Li: Ed, you wouldn’t have asked me that question if it wasn’t for Hollywood’s fascination with robots over the last three decades because if it was a law enforcement vehicle, a CCTV camera, or a fence, you wouldn’t ask me that question — like, “Hey, we’re going to break the fence or put shaving cream over the police motorcycles,” whatever. It would be a nonsensical discussion, why are you asking me that? And I think that’s the problem with robots in some regard is that these are tools, they’re not much more than that. And I think society’s expectation of what robots do is up here somewhere and reality is down here somewhere, the two shall meet at some point. But sure, it can happen. Will it happen? Sure. Has it happened? Of course. We have now put probably half a dozen people behind bars because, in all 50 states, it’s likely a felony for you to be messing around with a robot — no different than you deciding to go spray paint or knock over a police motorcycle like you probably shouldn’t be doing that. You’re going to soon realize that if you’re sitting there on three felony counts in the state of Washington for messing with one of the robots, people will start realizing, “Yeah, maybe I shouldn’t do that.” Remember, we likely have all the evidence to prosecute to the fullest extent of the law; we have and will continue to do so.

Ed Bernardon: Yeah, it’s interesting. You almost can think of it, at least from the law, from a legal standpoint that you have to treat it like you treat any police officer; don’t go putting shaving cream on a police officer’s eyeglasses.

William Santana Li: Property damage is the same thing, same concept. 

Ed Bernardon: The K5 versus the K3 — the K5 is an outdoor robot, it’s a little bit bigger. So, when you went from making one of these for the indoor operation to outdoor operation, was it a matter of just making sure it was waterproof, or was there other things you had to do to make it work?

William Santana Li: We went the other way. We did the harder one first, the outdoor one first, and then the indoor one was much easier.

Ed Bernardon: Yeah, if I can live outdoors, then you’re good to go. When you collect all this data, just like an autonomous car, you collect it, and now you’ve got to figure out, “Oh, is this a problem; this person laying on the ground? Or is this the same person that was there two days ago and four days ago?” How does the AI do that? In fact, it’s for sure different than an autonomous car because it’s not about what’s happening around you at that particular moment, but what’s happened over a long period of time, possibly.

William Santana Li: I need to adjust your question a little bit because you said “the data.” There are different flavors of data: there’s navigation data, meaning LIDAR scans and that sort of thing; there is data that might be associated with the health of the machine — conductivity, battery cell temperatures, and that sort of thing; and then there’s data that might be associated with the client’s security profile. So, not all data is treated the same; some resides on the machine and gets deleted because it’s just too much and we can’t use it. I don’t think we need the same LIDAR scan over and over again. There’s data that, for the machine’s health, we might want to plot over time, might be interesting. Then there’s data that sometimes will just reside on the machine unless the client needs it, and then we can write over it. There’s data that might go up to the cloud, get processed, and come back down. There’s data that the client might have asked for, it’s uploaded to the cloud, and then stays there and doesn’t get overwritten. So, there are all kinds of different flavors here.

Ed Bernardon: Like you said, there’s some data that’s for navigation just like an autonomous car; “Oh, there’s a bouncing ball or a little child crossing the street.” But you’re also doing “Oh, this is a person; they’re standing or they’re laying down or they’re running.” You do facial recognition or you recognize license plates. I mean, that’s a whole other level of rationalization; you have to go through that data, unlike an autonomous car.

William Santana Li: Yeah, and some of it, the autonomous folks probably don’t care. So, on the license plates, for example, if you pull up to an airport and you go to park in the parking lot, you pull up to the gate, the lights come flashing on, they snap a beautiful picture of that license plate, perfect lighting conditions, camera is not moving, vehicle is not moving, foregrounds not moving, backgrounds not moving. Tell me that’s a license plate, do the optical character recognition, all good to go. Simple, easy peasy. On K5, the machine is moving, the vehicle is moving, meaning the plate is moving, the camera is moving, the foreground is moving, the background is moving, the angle is just a little bit off, it’s 4:30 in the morning, it’s a little bit foggy. Tell me that’s a license plate and do the optical character recognition on it, and what does the plate say? That’s a little bit harder than just taking a snapshot at the entry gate for a parking structure. So, that requires a different level of machine learning and a lot of data. And then, in some cases, a lot of license plates aren’t exactly the same across different states. You got the reflectivity you need to work through, the speed of the camera, it’s all kinds of stuff. So, yeah, there’s a lot of headache. 

Ed Bernardon: Well, let’s maybe take it one level above that and let’s say, hypothetically speaking, that you know there’s someone from a particular state — I’ll just pick Colorado. Somebody from Colorado could be a threat for us for some reason. So, I want my Knightscope security robots to keep a special eye out for those Colorado plates. Is that a level of customization that you can build in? 

William Santana Li: That’s a feature that’s in production, and you wouldn’t call us up, you just literally do it yourself on the KSOC. So, on the Knightscope Security Operations Center, you can upload that particular license plate, and the machines then are on the lookout immediately. You don’t have to call us, that’s already in production and doable today. So, similarly, let’s say you are a major corporation, and you fired someone last week, and you’re worried that he or she might come back. You can upload the profile pic of that person, the license plate, and any and all mobile devices associated with that person, and now all machines are literally on the lookout for that BOLO or Be On The Lookout. And once you get that alert, that can be a notification on the user interface, it could be an email, it could be a text or phone call; again, back to the eyes, ears, and voice on the ground, 24/7, multiple locations at the same time.

Ed Bernardon: So, the person that’s buying your service, to some extent, really knows the reason behind the security. All you know, all your robots know is they’re looking for, as we say, a Colorado plate or whatever, it might be a certain face or something like that.

William Santana Li: Correct. A lot of usage in the casinos and that sort of thing. Again, that data is owned by our clients, not by us. They can upload like, “Hey, I threw this person out last week, this one’s card counting, this one’s dealing drugs,” whatever it might be, that’s on them. If the client is a law enforcement agency, then they have access to the NCIC database, then they can focus on stolen cars, felons, sexual predators, and that sort of thing.

Ed Bernardon: Before I let you go, let’s talk a little bit about the future. You got the K7 coming out at some point, can you tell us a little bit about the K7? What it’s going to do? How it’s going to be different than the robots that you have out there already?

William Santana Li: I think a couple of things to think about. That one’s a little bit more complicated than the machines that we’ve got. It’s the size of a small car and it’s intended to handle slightly more difficult terrain, not off-roading. 

Ed Bernardon: And it looks like a little car. 

William Santana Li: Some of our clients have gravel, dirt, sand, grass, border patrol, that sort of thing. But the sensor suite and the autonomous navigation get a little bit more complicated at that size and speed, so it’s taking a little bit more time. I think one concern that I have and then one opportunity. I think the concern is a lot of folks are cranky that we haven’t shipped the K7 out yet. I think you know this. Over $100 billion dollars has been invested in self-driving autonomous technology. There are, I think, 200 companies working on it. Their collective revenue is close to zero because no one’s shipped anything. Why? Because it’s kind of hard. And for us, why I get excited about K7 is we have constrained boundary conditions that an engineer can actually solve the problem, as opposed to random environment, random time of day, random everything. I need this 4000-pound unmanned vehicle to operate 100% successfully with no regulatory framework, no legal framework, or any technical specification whatsoever. That, to me, is a dumb engineering assignment. And I think that’s why it has taken and will continue to take a long time to get self-driving autonomous technology at scale, operating across the country. So, I think, the K7 has a lot of potential but we don’t want to release that until we’re ready-ready.

Ed Bernardon: Like you said, I don’t think any autonomous car companies are making money yet, much less even selling anything. Are you in the green? Are you making money with your robots? You’re certainly getting paid for your services, I would imagine, wherever you put these robots.

William Santana Li: We’re a publicly traded company here, so I’ve got to use numbers that are already released. We’re well over $15 million of lifetime revenue and we hold contracts, again, from Hawaii through Texas to North Carolina with dozens of clients across the nation, and we’re generating revenue, and continuing to grow. If you check out our social media profiles on LinkedIn, Twitter, Facebook, and Instagram, we’re announcing some growth news pretty much every single week like clockwork.

Ed Bernardon: Do you think that police forces will ever hire your robots saying, “We need more officers. Let’s hire a couple of robots.” 

William Santana Li: They already have. 

Ed Bernardon: So, that’s already happened. That’s not even in the future, that’s happening now. 

William Santana Li: Yeah. For example, one law enforcement agency three years ago said, “I don’t have an extra eight officers to patrol this park. We’re having all kinds of issues. We’re going to hire Knightscope. And not only better secure the park for the community, but we can have our officers focus on the rest of the city.” It went so well that — we didn’t ask them to do this — they went off and did the crime stats before and after. You can go to knightscope.com/crime and check out the details. But effectively, it kind of cut crime in half for them. And then what ended up happening was most fascinating. The chief of police then went before the city council requesting a renewal of said machine for another couple of years. And not only did he get the renewal, he got the renewal on a unanimous positive vote. So, I think that speaks volumes about the viability of the technology and it’s going to continue to grow.

Ed Bernardon: Have you heard of Boston Dynamics? They make robots that do flips and this and that. So, I had Marc Theermann, their Chief Strategy Officer, on the podcast. I’m going to ask you the same question I asked him. His vision, by the way, was to have robots someday be a companion in our home, just like you want your robots to be companions to the security people, to the police officers. And the question I asked him: Tell me — and in your case, it’s security robots — what will it be like 50-100 years from now? What will it be like? What will these security robots look like? What will they be doing? Will it be a friendly Robocop?

William Santana Li: I’ll answer your question, but one little side comment. I’m not a roboticist, I’m an ex-car guy. So, from my view, one of the biggest issues with the robotics industry is roboticists. We, the robotics folks, aren’t focused on fixing a problem; we’re fascinated with the robot. Boston Dynamics — they’ve got some very, very cool technology, but the company is 30 years old and have yet to ship commercially viable products out there. And I think we really need to focus on the technology being an application to help humans and help society at large, need to stop the fascination with what can be. I think the other problem with the industry is a lack of capital going into the sector. Everyone’s like, “Oh, well, those robots are going to do this, that, and the other thing.” Ed, if you ever want to figure out what is happening or what’s going to happen, just follow the money. Over $100 billion goes into startups every year, a tiny proportion goes into anything related with robotics. And it’d be different if $50 billion out of the $100 billion was going into this field, and all kinds of cool stuff is going to show up — no different than back then why does Yahoo, Google, eBay, Amazon, all these folks exist is because billions and billions were invested. Google is the 13th search engine; it’s not like it was the first one. So, I think you need to follow the money. For the future, it’s a non-event. You today, Ed, would not be allowed to build a new building without a fire alarm or a smoke detector, it would be like a ludicrous conversation like “Why would you even contemplate doing that?” It’s the same thing with the security robots, it’s going to be “Of course, you have these. Why wouldn’t you have these?”

Ed Bernardon: What’s one look like though?

William Santana Li: I think the premise of your questions, I would probably disagree — it’s not one, it’s a massive portfolio of small, medium, large, extra large, extra extra large, and extra extra extra large to cover every nook and cranny both indoors and outdoors, to better secure the places like I keep saying that you live, study, work, and visit. It cannot possibly be one technology and one machine. It needs to be a portfolio,  a very large suite.

Ed Bernardon: Yeah, so it’s not necessarily the humanoid robots running around replacing police, and there could be stationary sensors. Who knows? Just an array.

William Santana Li: The humanoid thing is very cool on TV, but until you’ve got energy solutions/battery solution and a lot more efficient in electromechanical stuff and computing, it’s just not worth the time.

Ed Bernardon: You should listen to the podcast with Marc Theermann from Boston Dynamics because one of his goals is exactly what your goal is: to take the technology they have and make it practical. And they do have now warehouse robots and Spot, that little yellow dog, is actually used in factories and things like that.

William Santana Li: I’m hoping they make more progress; they’ve got a lot of resources there. We certainly wish them well. It’s just that, to me, we’re still in the equivalent of the 1970s of the robotics industry; it’s similar to the PC industry, I should say, in the 1970s. It’s just coming out of the hobbyist category. There are a couple of companies here and there making a little bit of progress, but not at scale. And we need to pick up the pace here.

Ed Bernardon: We talked a little bit before about the Roomba. So, my final question for you, at least from a technical standpoint, is when do you think I could buy one of these from you, have it walk around my house, keep an eye on my pets, probably not watching my kids but let’s say watch my pets, do security for my home, make sure the faucet’s not leaking too much? When do you think I could just buy one of those? And I almost treat it like it’s– I’ll name it and it’ll be part of the family. Could it become affordable?

William Santana Li: That’s not on my to-do list. Do you have a human person to do that today for you? Do you pay? One of the criteria for us to work with a client is like, “Do you have a security solution today? Are you paying for that today?” That’s a Jetsons cartoon, a nice-to-have, cool thing in the future, but are you willing to actually have that in your home? In the short to medium term, I don’t think that’s a viable thing. It would be different if you were to ask me, “How do I get my HOA or my neighborhood to band together to better secure the neighborhood? We would be willing to pay for that. We don’t pay a police officer or a security guard to do that today, but the neighborhood wants to do something.” That’s a lot more viable, and that could happen this year as opposed to the next decade. But it’s the “I want to have, it looks cool” kind of thing, but the practicalities of it, this is back to the fascination with things as opposed to fixing an actual problem that you have. Are you really checking every day that the water is leaking or not in your faucet that’s like a paramount thing that’s really critical to you? I’m not sure that you’re willing to pay for that. 

Ed Bernardon: Bill, thank you so much for opening up our eyes to a whole new area where robots are used. As the last part, traditional, on the Future Car podcast, we’re going to do our rapid fire. So, real quick. I just got a bunch of questions, quick answers. What was the first car you ever bought or owned?

William Santana Li: 1974 Camaro Berlinetta.

Ed Bernardon: Did you pass your driver’s test on the first try? 

William Santana Li: Of course. 

Ed Bernardon: Tell us have you ever gotten a speeding ticket?

William Santana Li: I used to have a radar in my car, I took it out for obvious reasons. I lived all over Detroit, but when I used from Ann Arbor to get to Dearborn, and it used to snow when I was typically driving a Mustang GT. And it was that good to be driving sideways in the snow to work every day or treating it like the Autobahn coming back home. So, I had to get rid of the radar detector.

Ed Bernardon: You like that drifting feel then?

William Santana Li: Oh, yeah, driving sideways.

Ed Bernardon: We talk about autonomous cars, and we like to think of them as like a living room on wheels. Five-hour trip, you’re in your living room on wheels. You don’t have to drive, so you can do anything you want. Describe the inside of your living room on wheels as it takes you from Detroit to St. Louis.

William Santana Li: I’m probably working, Ed, I got stuff to do. We got a country to secure, so I’m probably figuring out how to land a new client or hire a new person or do something else.

Ed Bernardon: So, you have an office on wheels, not a living room on wheels. What person, living or not, would you want to have with you on that five-hour ride, to help you do all the work you need to do?

William Santana Li: Help me do the work or would I want, that’s two different questions, Ed.

Ed Bernardon: Oh, whatever you want. In fact, they might distract you from the work, that might be better, then you can actually turn it into a living room.

William Santana Li: I won’t name a particular person but a type of person. So, to me, founders of companies typically have a screw loose, which I think is awesome because you’re required to do that. And most folks don’t realize what it actually takes to build a company, much less start a company, get it funded, grow it, and take it public. So, I always find conversations with other founders fascinating, invigorating, depressing, and motivating, all at the same time.

Ed Bernardon: Which one would you pick then? What founder? All right, you can have two, I’ll let you have two. Usually only one, but you get two.

William Santana Li: We’ll put Steve Jobs and Elon Musk in the same vehicle together and we’ll have it all out.

Ed Bernardon: Yeah, you could just sit back and let them go at it. If you could uninvent one thing, what would it be? 

William Santana Li: Facebook. 

Ed Bernardon: And if you could magically invent anything?

William Santana Li: Something that properly deploys risk capital to startups. The fundraising process for building a company is broken, and the entire VC industry needs to click the undo button.

Ed Bernardon: What Knightscope robot best describes your personality? Of all your models, which one is The Bill Robot?

William Santana Li: No one’s ever asked me that. My personal pet project–

Ed Bernardon: Yeah, who’s your favorite? I hope the other ones aren’t listening. 

William Santana Li: I got some other stuff up my sleeve, but what I can publicly disclose, I would say the K7. I’m still a car guy, I want that vehicle out the door.

Ed Bernardon: Last question: tell us something about yourself that would surprise your friends and family.

William Santana Li: My friends and family. The newer folks that didn’t know me back in the day, I used to have really long hair and play guitar in a heavy metal band. So, I’m looking to pick the guitar back up.

Ed Bernardon: That explains that Van Halen guitar in the back.

William Santana Li: Oh, yeah, I got a Marshall Stack over here and the new EVH 5150 head.

Ed Bernardon: So, we’re going to create some beautiful artwork to help people understand what’s on the podcast. Please send us a photo of you in one of those hard rock bands with your heavy-metal-band long hair and the Van Halen guitar. And we’ll show the transition. Is that a deal?

William Santana Li: I’ll get you something. Let me go pry through the archives and figure out something.

Ed Bernardon: Let’s see what you can find. Bill, thank you so much for joining us on the Future Car podcast. It was a pleasure to talk to you.

William Santana Li: Appreciate all the questions. Thanks for having us. Please be safe out there.