Becoming a Metapilot

Tullio Siragusa (00:12):

Hi everyone. Good day. We have to say good day because I don't know where you are. It might, it might be morning, afternoon, or even evening. Our guest is all the way broadcasting from Dubai. This is Tullio Serosa with Tech Leaders Unplugged. I am getting unplugged today with Diana Deca. We're going to be talking about becoming a meta pilot. That's right. A meta pilot. And the theme today is bringing Flying cars to Life through VR Gaming. So this is a very interesting conversation we're going to have, and I want to welcome Deanna to the show. Welcome, Deanna. It's good to have you.

Diana Deca (00:49):

Hello Tullio. Thank you for inviting me. And yeah, very excited to start chatting about this. I think it's a really fun topic not just because I was a gamer myself growing up, but I think we're living very interesting times. So yeah, looking forward to this.

Tullio Siragusa (01:05):

Very much so. So let's just dig right into it. First of all, I'm curious to learn how this whole idea came about. You kind of hinted it a little bit, being a gamer or, or a next gamer, or maybe you're still a gamer. Tell us a little bit about this idea. You know, what was that moment where you were sort of like, I want to go build this thing and, and, and please share that with us.

Diana Deca (01:28):

Sure. I mean, it will seem a bit intricate the way I do it. So it started, of course, in childhood. I was a nineties kid and I loved you know, what we call now retro gaming <laugh>, which sounds bad. So I was playing, you know, Wolfenstein Quake Doom, all of these games, right? And it was very exciting at the time. In the meantime, I ended up doing a Ph.D. in, microbiology in Munich. And in there my project was about getting some mice or some, you know, the actual animals, the mice building a little VR arena for them. And so the mice were running on a ball and looking at the VR screen. And in the meantime, I was sticking electrodes in their hippocampus and trying to understand if we can convince the mouse brain that the VR is real.

Diana Deca (02:19):

That question fascinated me at the time because that means, you know, it means we can erase the boundaries between what is real and what is assimilation. So that was a very long time ago, long before the metaverse appearing as a buzzword. Long story short is very complicated to do that. Our brains are, are very advanced in that sense. But yeah, it's, it's been very interesting. And so the game meta pilot is very much about that. It's, it's about bridging two worlds, the digital and the gaming universe with the existing one. The main thing about the game, though, is it's actually designed with a purpose. The flying cars that you experience in the game are real. They are in development and about to be deployed at scale. So by playing the game, you're actually generating synthetic data that helps these companies go to market faster. It helps them launch in more cities faster. And so, yeah, it's, it's partially because I'm nostalgic from the nineties and the eighties, you know, if we watched the Jetsons or read about Isak, ASIMO flying cars was a dream then. And it is, you know I wanted to stop being the dream today. I hope it happens very soon.

Tullio Siragusa (03:29):

Very interesting. Some of us started with, you know, pong on Atari, but nonetheless, we've come a long way. You said something really interesting that caught my attention. The folks that are using the game are actually piloting real cars that are in production. So in essence, you're providing, you're gamifying the process and dare even say, there's probably some design thinking in there because you're actually connecting with the end user before you release a product. Tell us a little bit more about that partnership. How's that working?

Diana Deca (04:07):

Yeah, so in, in the first stages, the purpose of the game is to generate data, right? So at some point, I'm probably going to show you, but in Meta pilot, we now have a very simple version of Manhattan. So, you know, if, if you're from there, you can probably recognize your building. And we have a few simple air taxi models. So as you're playing the game, you're generating data about your flight and how you actually flew the aircraft a bit like an Uber, which is exactly how they're going to be in the next two to five years. But we also get a lot of other kinds of data. For example, we can simulate different weather events, which is something that regulators are concerned about, right? So if you were to go into a flying car right now you might be a bit scared, you know, knowing that there's a storm and you're flying right next to a skyscraper, right?

Diana Deca (04:55):

So we can actually get that data through you and from you. And, and the best part about this is the diversity of the players, right? You can have former Boeing pilots playing the game. You can have, you know, top guns, but then you can also have 12 year old kids or students who are excited about it and who don't have, sorry, very annoying fly <laugh>. You can also have, you know, people from wherever in the world that didn't get a chance to, you know, go to a flight school or, you know, drive a very expensive aircraft. And through the game we can actually find them and, and locate the most talented future pilots.

Tullio Siragusa (05:31):

Interesting. Nothing like being live, right? With flies flying in front of us,

Diana Deca (05:35):

<Laugh> I think it's on purpose.

Tullio Siragusa (05:37):

So, so I'm, this is I'm curious about the applications, the broader application of this, right? So I see the value in basically giving the feedback loop for companies r and d. I mean, they've gamified a way to do that and it's fun for people to participate. Are those people participating as a subscriber or they're getting paid for this kind of a, playing the game? How's that set up? What's the, this model look like?

Diana Deca (06:08):

Yeah, so again, since we're in the starting stages, we've found that a lot of people just want to play the game. I don't know if it's the movie Top Gun or it's something else, but it is very cool to actually play it. The game is very accessible. It's in the range of 15 to $20. Again, we might do a giveaway after this. And then we also work with the companies to upload their air taxis there. But since you asked in the future, I mentioned before that, you know, we're going to have a lot of different kinds of people playing the game. We're going to have 12 year old kids, and then we're going to have Boeing fighter jet pilots or even people who just finished aviation school and are ready to become pilots. So for the second group, imagine if you were, you know, from Boeing or Airbus and wanted to test a specific type of aircraft, you could create missions and these people could get paid for what they, the data that they generate. So in that sense, it can, well, it has the potential hopefully to become a sort of an Uber.

Tullio Siragusa (07:10):

Well, I mean, I keep, I keep thinking about that movie, the fifth element. You know, it reminds me of all those flying cars and taxis and and so on. But, but I find the business application of this really intriguing, and I think a lot of people have struggled to create value for the Metaverse, for example, in a real way. And so what I've heard so far is I can actually hire your company if I'm producing a flying vehicle or next generation air airplane or whatever it is, and I can hire you guys to go and basically get a bunch of users to test my product out to virtually put it through its paces and give me feedback. And I could generate through that feedback, all kinds of data and input that could help me with fine tuning my production. My r and d I'm wondering what other uses could this be applied to? You know, there's so much that this can be worked on, even for, it could be used for eventually surgeons having, you know, virtual practice using this kind of technology. What are your thoughts? What's the future? What is this going to open up for from a broader perspective in your, in your opinion?

Diana Deca (08:21):

Yeah, great question. So, well, the reason why we started with air taxis is because, you know, they're a great beachhead. They are an emerging market, you know, they're much easier to deal with and for example, BMW or Tesla, right? Which have their own internal huge r and d teams. But in the future, this applies to many things that are in development and are really willing to do a lot of back and forth with us to understand their market. One of these applications is governments and defense, right? So because we are quite neutral and we work with all of these companies, we also function as a platform for governments to be able to test different types of air taxis. So if you want to see, for example, with Urban Aeronautics, which I believe is the series C air taxi company they were looking at deploying their model as a flying ambulance, right?

Diana Deca (09:11):

Because it can land vertically and it can take off vertically. We were able to make an estimation of how many lives per day they would save as a flying ambulance. And so that, that makes a great point for pitching this to a specific government or a specific city and so on and so forth. So air taxis defense, but it can be pretty much any type of vehicle in the future. And even as you said, I mean, the possibilities are endless, of course medical applications, I think even for us, it's actually quite hard to to fully grasp where this is going to lead us. But I think since you're into movies, maybe considered the movie Tron

Tullio Siragusa (09:48):

Yeah, that's

Diana Deca (09:48):

A good example.

Tullio Siragusa (09:49):

I mean, it's, it's interesting the, the, the way this is being applied in a real way, you know, you're marrying the real world with the virtual world to provide feedback loops so that you can actually create better products in the real world. So now let's talk a little bit about some use cases that you foresee. I you talked about the taxi use case which sounds phenomenal. I'm wondering now, do you actually need to have people piloting these taxis? Or can it all be done remotely? What's the end goal? You know, I, i, I foresee the possibility where I can have a, a, a bank of people operating remotely who can, you know, who can move things, be at different places at different times. I mean, the boundaries that exist in the physical world. I e I have one car and I I'm in one location and I'm one driver versus a remote driver. And I can be in many locations, in many cars very, you know, pretty much instantly from one to the other. What's the thinking there? Is that, is this more of a r and d thing where you then have the, the real drivers? Or is this the ability to actually do remote operation on these kinds of vehicles?

Diana Deca (11:04):

Hmm, yeah, I mean, this is an overarching discussion, right? And similar discussions are happening around chat G P T or any other similar methods especially with the metaverse. So what we see right now, I, we can easily compare to the case of autonomous cars or autopilot, and we can look at, for example, Tesla, right? So in the case of Tesla, as you saw, they actually collect sensor data and location data from the drivers themselves. If you sign a contract with them, your car will be transmitting that at all times. This then goes into their database and ultimately trains their AI algorithms, which become better and better at driving autonomously, right? And then, you know, in the future, in the foreseeable future, their claim is that they will become fully autonomous. It's sort of a similar approach where the algorithms learn from the human data and then humans still, you know, continue to interact with the algorithm and also learn themselves.

Diana Deca (12:03):

Except that this now is much more interesting because it's not just driving a car at 2d, it is driving in 3d. So the algorithms can be trained in many ways and considered defense and strategy. Imagine being able to simulate the same event a few hundred thousand times rather than, you know, doing it cowboy style and seeing how it goes. Right. So raising the success rate of something like that to 98, 90 9% is, is very interesting. And it's also very interesting for space applications. It, it prepares us also for different gravitational situations where we don't know exactly what, you know, the gravitational pool is, or we don't know how the weather is going to be like on Mars, or how dark it will be, or what other flying objects there may be around. So this is a great platform to train for all of the unknown unknowns, let's say.

Tullio Siragusa (12:52):

Yeah, that opens up another thought process in terms of safety improvements. I I, I'm thinking like, imagine having someone like Maverick on Top Gun flying a vehicle, and basically the machines are learning all these various tactics that can be incorporated. Is the, the thinking that some of the input from the r and d could also become training modules for the ai, so that if, if I'm an operator of a flying vehicle and I get into a problem area, the system takes over because there's an expert driver or someone who's, who's got more experience than I do in flying that vehicle that, that it takes over the operation and sort of has that module bill thing. Is that part of the process you guys are thinking about in terms of enabling safety standards? Because that's a big concern for a lot of people, you know, is are these things going to be safe?

Diana Deca (13:46):

Yeah, absolutely. So that's actually why we started the whole thing long before we built the game. We worked very closely with regulators at Boeing, at nasa. That's why we also got some funding from Boeing and their partners to be able to generate data that is actually interesting for regulators. And we chair a few of these committees where we're trying to see, well, what kind of data would be used, how can we combine it? And our thinking is, for example, a new emerging air taxi concept appearing in New York, right? It would be great to have, you know, a hundred flights in Manhattan, but what if you can get a few hundreds of thousands of flights from people and then simulate those for all of the, again, unknown unknowns, right? And then simulate bad weather, simulate storms, or whatever else there may be happening. So we're trying to combine all of this and increase the safety levels also keeping in mind, and that's something that we get into quite often.

Diana Deca (14:42):

A lot of people when they think of air taxis, they think of the jet sense, as you said, or the fifth element, right? And that's a lot of pressure to put. It's a bit like putting pressure on on a kid, right? Like on a, on a teenager. So at this point we might want to see air taxis as slightly cooler helicopters, okay? There are helicopters moving about all the time. Slowly they will become a bit more autonomous, let's say level three or level four like cars are right now. And then slowly we will add things to them. For example, we will make them fully electric. We will make them hydrogen based. We can create an AI system that controls them and make sure that no accidents happen and then, you know, sends them to Mars or whatever. So we can do this in a very incremental manner, and that's why it's really important to talk to people and also talk to the younger generation and make sure we check in with them and not just, you know, project these sci-fi dreams onto them, but rather let them get involved.

Tullio Siragusa (15:39):

All right? So we're not going to be getting flying cars anytime soon for the regular person, but it's really commercial applications is what I'm hearing. What are some of the challenges that you faced in making all this work? I mean, this has gotta be very technology intensive. Were you able to find some open source tools or did you have to build proprietary technology? How did you go about putting it together?

Diana Deca (16:07):

Yeah, good point. So, well, I did quite a bit of this with my mice where we kind of had to start from scratch because you don't have a lot of funding in academia, right? So it was good training for being a startup. I mean, you know, unity and real engine, they're, they're developing really fast. They're really great. I think the trick here is to a, make sure you generate data that is interesting for regulators and government. So that's something to do before starting the game. And then second of course you know, building a pipeline that sends that data somehow into your servers. But finally, the most important part is the backend, which is the AI backend, which generates that data and analyzes it. And this now is based very much on my Ph.D. research. So in a sense, we're treating the gamer, the meta pilot, very much like we were looking at the mouse brain trying to extract as much data and with a lot of curiosity.

Tullio Siragusa (17:04):

Interesting. Okay. So it sounds very complex and challenging to put together and, and, and make it all work. So when we'll be able to see this, is this in production now? The

Diana Deca (17:20):

The game is available right now also in VR and it's available in several amusement parks actually, that we collaborate with. You know, I, I could do a demo right now. I could show you videos if you want, but yeah, you'll see that if you go through. How does it work?

Tullio Siragusa (17:36):

Require any hardware? I do I need a device or I a room set up for this? How is it set up?

Diana Deca (17:45):

You go on Steam, well, you go on our website and your body study eye, and then you go to become a meta pilot, it will take you through the game website and there you have the option of playing it by a desktop or you can play it, you know, you can connect your VR goggles to your laptop if you're into that, and then do the VR version in the future. There will be more complicated versions of this, but it's available. That was the hard part. Really. It's available right now. People have tested it, A lot of influencers have tested it, and we have improved it a lot based on their feedback.

Tullio Siragusa (18:15):

Yeah, I keep thinking about the simulations that NASA and even the d od uses for fighter jet pilots. You know, they put in time in the simulations to, to practice. Is, is the thinking for the more advanced r and d to actually have these kind of virtual meta studios, you know, or, or is that not necessary? How, how close is this to actually just being able to use your screen on your computer to, to provide real valuable input?

Diana Deca (18:47):

Right. yeah, so we collaborate with all of these groups with a lot of governments and, and we're working with NASA on this. So there are different levels, right? And that's, that, that was probably the hardest part, building a game that is actually engaging and relevant for people who are now, you know, 12 to 19 years old. And we did that by having, you know, people on our team that are within that age range and extremely talented usually coming from universities like UCLA and NYU. So there are levels, again, you can be slightly interested in the game and you can be very competitive about the game. You can follow our development and try to get involved and then just email us directly from the website. And in time there will be a path to being noticed by different groups, be it the air taxi companies, be it Boeing, et cetera. So that's, that's ultimately our goal. I would say if you were born in a random place in, in Ohio, let's say I, I took that completely randomly and are into this and don't have your own fighter jet to play on and no one can pay for your school, this is the way to get noticed. So that's

Tullio Siragusa (20:01):

Very interesting. And then that, that brings me to the next question I was thinking because you have essentially the potential for two business models. One is the gaming model itself, where people can subscribe and be, have access to something really cool. They're actually driving real pro prototype products that's in, that's in the real world. And then on the other end, you have this ability to provide incredible r and d and input and training modules and simulations for commercial applications. And it sounds as though you're sort of marrying these two where the entry point could be you're a player and you're having fun with it, but you know, there might be a need for future people that can contribute input and so on by these commercial entities. What's the end goal for you? Is it this hybrid of these two, or is it predominantly, you know, a game or is it predominantly focused on the r and d input? What are your thoughts?

Diana Deca (20:57):

Yeah, great question. So how, how to bind the two for now? Indeed, they are two revenue streams. They're both moving fast enough, but going back and now it's getting full circle. I like this. I did mention that we'll have different types of players. Some of them will be very young, potentially very talented, and some of them will be well-known pilots or people who are taking a break or people who are actually, you know, trying to get back to their normal life after being a fighter jet pilot. That's another topic for another day. So, the latter, the second category is more likely to become connected with those companies, right? And generate valuable data. But in time we are also creating a path for everyone else to, you know, get involved to an ex to the extent that they want to get involved, really. And to have that freedom to decipher how much they want to know about how deep the rabbit hole goes.

Tullio Siragusa (21:52):

Very interesting. So what are some of them, is this your first entrepreneurial venture, Deanna?

Diana Deca (22:00):

I was involved with actually two startups before, and new robotics has evolved. It has had different applications, mainly for an AI company. We've approached different things from medicine to manufacturing, and we still have those connections. But when it comes to something that scales at the pace that our team wants it to scale, I think meta pilot is a great way to connect people and technology. So yeah.

Tullio Siragusa (22:25):

So what's that experience been like? We're coming up to the end here. I'm just curious. I mean, this is a very complex, interesting topic. It's been met with a lot of support by some groups. It's been met with a lot of pushback by others, you know, people that have dismissed the metaverse and blockchain. There are always those who just, just do that because maybe they don't understand it. What's been like for you building this thing?

Diana Deca (22:50):

It's been very interesting. I mean, it's, it's really the core of what I was doing. So I was happy doing this in my Ph.D., you know, 12 hours a day with no weekends. So, you know, it's been actually what's behind your robotics has been going on for 15 years, probably. As for the expert the startup experience itself, which I highly recommend to more scientists is very interesting because it really humbles you. Academia has a way of building oversized egos, I think, which i, I am also guilty of. And the process has been amazing because I learned so much, I learned so much from people working at NASA, from fighter jet pilots, from angel investors, from VC funds you know, and all the corporations and partners that we work with. And I think it's, it's just really cool for me. And I'm very grateful to have studied neuroscience and still be able to chat with fighter jet pilots. I mean, what, what more could you ask for?

Tullio Siragusa (23:47):

That's amazing. I admire that you've made this your lifelong mission. So it sounds like we're going to see some amazing outcomes from this. So anyone who's curious about this, they should check out your website, new robotics, and if we can post that, there it is, robotics.ai. We wish you tremendous success. So we're up on time. So what have we learned today? We've learned that you can marry the real world with virtual reality and you can create value for both people interested in just leveraging and consuming fun things in the virtual world, in the metaverse, and also provide valuable input to the real world scenarios. And here's an example of those two coming together. Who knows what this will open up? There are so many other potential opportunities to create this kind of marriage if you will. So we'll, we'll, we'll see how it all plays out.

Tullio Siragusa (24:44):

AI is making a lot of things possible that we, we haven't even thought of in the past except for sci-fi movies. Deanna, it's been a pleasure to have you with me this morning this afternoon or for you this evening, wherever you're watching. Please stay with me as we go off there in just a second. And we've got a lineup of other guests this week. Tomorrow we are going to talk with Joel Neeb who's VP of Execution and Transformation at VMware. We're going to be talking about From Reactive to Proactive, how to transform IT operations with AI, and eliminate operational debt. So that's an interesting conversation. Anyone who knows anything about technical debt or operational debt, should definitely tune in and watch that show. It's going to be, we're going to try to go as deep as possible. And then I think we have another guest coming up later this week. I could be wrong, but I, that's the one coming up tomorrow. If we have another one, we will announce it tomorrow. Thank you. Oh yeah, we have one more on Thursday. What do we got coming up on Thursday? So, we got three shows this week on Tech Leaders Unplugged. We get to get unplugged three times, so tune in tomorrow same time, and we'll announce what's coming up the day after. We have Bob Rogers, who's the CEO of Oii. So thanks for being with me. See you again tomorrow. Same time.