Bob “Sully” Sullivan: Lior Tal is the CEO of a company called Cyngn - C-Y-N-G-N. And they are publicly traded under the stock symbol $CYN.
Why aren't you, why aren't your industrial vehicles driving themselves? Is the question on their website. Pretty interesting stuff. We're talking about the shift to advanced autonomous vehicles. And I think this is something that was a dream ten years ago, maybe five years ago, and it's a reality now.
First of all, welcome to the program, Lior, tell us about what you guys are up to, because this is interesting stuff. This is like Orwellian almost. You have autonomous vehicles happening, only you guys are actually pulling the trigger on it, it’s actually happening.
Lior Tal: First, thanks for having me here. This is great.
Bob “Sully” Sullivan: You're coming back already. I can tell because the topic you're coming back over and over again.You can’t get out of this.
Lior Tal: After all those zoom meetings being in and out, it’s great.
You know usually when I explain to people the difference between automation and autonomy, if you take it a little bit to an extreme, it's like the difference between the Thunder Mountain on Disney and what Tesla does on the highway.
Bob “Sully” Sullivan: Sure.
Lior Tal: Both don't have drivers, right. Just one response in real time to things happening on the road. The other just repeats a very structured environment.
So what we do to the industrial world is what Tesla does to the highways. We can bring that level of driving into these industrial applications, whether it's construction, mining, warehouse logistics. And we can do it by retrofitting their existing fleets, which is a very big difference because there are millions of these vehicles already out there, and they don't really want to throw them away and go buy a new robot that does what it does before.
So it's a unique approach. We've been working on this since 2016 and took the company on NASDAQ at 2021. And we're here now.
Bob “Sully” Sullivan: Where are we at in the shift to autonomous vehicles? Where's the arc of that story? And where were we, let's say, because I've heard this talked about ten years ago and five years ago, but you never saw anything, and now you're starting to see this. Are we starting to crack that surface, or has it been that way for a few years?
Lior Tal: So I think something very interesting happened. When people think about AI and robots, they think Skynet, they think you're the robots are coming to take us as.
Bob “Sully” Sullivan: As you just brought up in Terminator. Perfect. Exactly right.
Lior Tal: It’s true. But actually these are just very sophisticated tools, right? And there's an evolution of the technologies that allow to build these tools.
Sometimes you need a factory to build a car, right? And those factories started evolving around the beginning of 2000’s, at the DARPA challenge. And what that really started is a really renaissance of developing components like laser scanners, like AI software infrastructures that allows companies like us to come in and really just build what's missing. The driving software, the management software.
Where companies like Nvidia and Google and the others did, they have a lifting to build the infrastructure that allows us to do that.
So going back to your question, we are much closer than ever before to actually being able to productize, commercialize these technologies. It's still years away on the roads because of the complexity, the speed and the regulatory space. But it's absolutely now inside the industrial world.
Bob “Sully” Sullivan: What kind of challenges with respect to regulation are you seeing? I mean, the first thing I think about is OSHA. The first thing I think about is suddenly there's a costco, there's a big forklift that suddenly can be robotic thanks to Cyngn, ok. And suddenly they're more efficient, they're faster, they're saving money. I don't know if they're necessarily taking jobs or they're certainly augmenting jobs.
But then to start to think about, when I saw the video there, there's a robot going around, a worker going around the corner and stuff like that. Any challenges with regulatory bodies in terms of safety?
Lior Tal: Let me maybe start talking about the safety aspect and go back to regulation. Right? There's a reason we're asked to count 100, 102 when we follow a car, right? Because our processing time, our response time as humans is relatively slow. It takes us about a second and a half from the minute we sense the world, to make a decision to actually pull the brake. Right? And we actually only see in a very narrow field of view.
Bob “Sully” Sullivan: Sure. Right.
Lior Tal: Now, in contrast, our machines, they have 360 degree vision, cameras, lasers, radars, whatever you need. They don't play with their phone, they're never distracted, they can calculate…
Bob “Sully” Sullivan: This is the same argument why auto driving cars will not have as many accidents as you and I tried to drive down the road. Correct?
Lior Tal: Exactly. But in the industrial world, it's there because it's 5 mph. And on the highway, it's 60 mph. Right? In the industrial world, it's fairly structured, and on the road it's an absolute chaotic environment. So the long term is different.
So the point is that inherently these machines are safer because they see faster, they see more, they decide and act faster. And I think, like everything else, the regulator, they're smart people, they catch up, they need to see the statistics, they need to understand the implication of using it. But it makes a lot of sense because eventually one of the problems that autonomy solved in the industrial world is safety.
Bob “Sully” Sullivan: Yeah.
Lior Tal: There are a lot of accidents.
Bob “Sully” Sullivan: And is that what's happening they are sensing versus you guys preprogramming? Or is it a little bit of both?
Lior Tal: So if you pop up the hood, what's really happening here is it mimics the way a human would drive. So the sensing is the raw data coming. With us, through the eyes and ears and smell. With the machines, it's usually lasers and cameras. Now that data comes in and then gets processed into milliseconds.
Bob “Sully” Sullivan: In milliseconds correct? Yeah.
Lior Tal: Yeah Lidars work at 10 times per second. The whole thing repeats itself, and what happens is that you can teach it to perceive certain things that are relevant to the environment. I can teach it what a forklift is or what a person is or things like that.
And you can evolve it. It's like the scene from The Matrix where I know kung fu. Right?
Bob “Sully” Sullivan: Right, sure.
Lior Tal: It's the same thing. You teach it, you push it down, and the vehicle knows it.
So what happens after? Now the software knows what's happening around it. It needs situational awareness, the context of what it needs to do and what's happening around. It makes a decision after calculating hundreds of different possible driving options, sends it to the vehicle, repeats the whole thing ten to 100 times per second.
So it's really the way you and I would drive, but just using very strong sensors and computers.
Bob “Sully” Sullivan: You and I would drive. Maybe not Huckleberry over here.
Lior Tal: But we don't judge.
Bob “Sully” Sullivan: Lior Tall is the CEO of Cyngn, Stock Symbol, CYN. CYN is the stock symbol. You can go to Cyngn, cyngn.com.
I got a question for you with respect to the trends in robotics. Okay. There seems to be less fear. Adaptability seems to be a juggernaut now, because I think of the efficiency model. Is that something that I'm assessing because I keep on reading the good side of robotics, or is this actually happening? Are you having less and less interference and trying to get in the door now? I'm assuming it's easier than it was.
Lior Tal: Oh, I'll be honest, when I started working on this, the big shiny object was cost of labor. If you look at these vehicles in the US, the labor cost to drive them is over $100 billion per year.
What's actually happening when we talk to customers, especially in the last two years, is they just can't hire enough people, and it takes time to train them, and they don't like the work. It's mundane, it's repeating, it's dangerous. So what actually we're seeing with customers is they're trying to use these technologies to augment the workforce, get more out of the cost of real estate they're paying, increase productivity, reduce the cost, increase the safety.
Now, once the conversation went from, “is it replacing my workers?” to, “can I leverage my work or upskill them, have them do more with these tools?”, the adoption is much faster.
Bob “Sully” Sullivan: Sure. Exactly. So you've got an increase in productivity and the savings in cost of labor all at the same time. I think that's it at the end of the day, right?
I want to talk about what's next for you guys, what's exciting you the most out of the next 18 months, in the next five years? Because that's obviously technology is part of it, but you know where the rubber meets the road here.
And as a CEO, you can tell us some stuff, but not everything. But what can you tell us about the next 18 months in the next five years, in your opinion?
Lior Tal: So, you know, when we went on a roadshow at the end of 2021, we told investors to look for certain milestones in 2022 to make sure that they understand we're on track.
The first one was steph up the team. We were 40 people when we IPOed, and it was post COVID, we needed to add more people, especially around engineering.
Second thing is we want to stay a software company and a data company. We'd like to work in collaboration with the vehicle manufacturers to go to the end customer. We don't build vehicles.
Bob “Sully” Sullivan: Sure.
Lior Tal: And the third thing is pilots. To show beginning to end how our software with their vehicles gets in the hands of customers.
And our plan was to take one vehicle, one customer pilot in 2022. You know, at the beginning of the year, when we started seeing what's happening with the economy, we decided to accelerate. The plan was just to be more aggressive and try to pull the time frame to revenues shorter.
So we ended up in 2022 actually working with two different vehicle manufacturers. One is the Columbia Vehicle Group for Stockchasers. The other is Greenland for forklifts. We've actually had a handful of deployments in manufacturing logistics.
What's coming is a larger scale, it’s a faster execution. So in ‘23 is when we expect revenues to start trickling in. These deployments are going to go from a single vehicle to dozens of vehicles. Everything leads to ‘24, ‘25 when we expect to really start going.
Bob “Sully” Sullivan: One last question for you, because you talked about one of the key selling propositions here is the fact that these manufacturing or these warehouses don't need to buy new vehicles. They don't need to buy new robots.
Will there be a time when you are OEM for some of the vehicles that are being sold in the marketplace? Is that on the table?
Lior Tal: No, it's a good question, because I've built several companies, and in every company you touch metal, you touch atoms. There is an argument at some point, “let's just do everything in house”, “let's just control everything”. I think that as long as we can, we would like to remain a software company by structure.
Bob “Sully” Sullivan: Sure.
Lior Tal: And I think it's just a more interesting business. Allows me to work with a long list of other vehicle companies. And really, it's a different skill set.
Bob “Sully” Sullivan: You got to come back, a lot. There's a whole show on it. Exactly. Lior Tal, CEO, Cyngn, Inc. C-Y-N-G-N their sock symbol is $CYN.
Damn, that is a head scratch.