Why XYO Network sees trustless data as critical for AI and robotics

0:00I'm Ashen Addison from the Crypto Coin Show and today on blockchain interviews we have back with us Marcus Leven, co-founder of XYO here to talk about critical infrastructure for AI and robotics, trustless data for autonomous systems, decentralized networks, and the continued growth of the XYO layer 1. Marcus, welcome back to the show and thanks for taking the time again.

0:25

Thank you, Ash. It's always great to be here with you and your community. Likewise, I'm pretty optimistic on 2026. I think 2025 was pretty good for XYO in launching the layer 1 network after achieving 10 million nodes around Q3. I'd love to hear how that's progressing now that we're a couple quarters in. And first of all, I'd love to get a recap for those who are who

0:51may have missed some of the previous episodes on a high level on X YO what the vision encompasses and then a little bit on the layer 1 network and then we can dive into all those topics I mentioned at the beginning.

Yeah. Cool. Awesome. Yeah, it was a busy time the last few few quarters for sure. Yeah. But let me start with Xio. So you know we are a reality engine

1:15connecting the real world with blockchain and we started doing that with something called proof of location with proving the location of devices or people and then from there we move to proving data like collecting data and then proving that it's true and correct and then connecting it with people on the web two and web three side and from there you know we grew our

1:41network of devices. We started with these IoT devices like Bluetooth beacons and GPS trackers and today we collect a lot of data through mobile phones and in other ways as well. That's useful for a number of industries you know if you think about smart cities and self-driving cars drones and you know the world of the future basically and as well as on the

2:06data side you know for AI companies and other types of companies which have needs for lots of data. Yeah, the AI is growing so fast and you know personal robots could be here by the end of the year or by next year for sure. According to Elon Musk, starting XYO back in 2018 as a decentralized physical infrastructure network, the landscape was completely

2:33different with, you know, AI and robotics and it was in the works, but it's growing so quickly h as the, you know, outside of the echo chamber of blockchain continues to grow, autonomous vehicles and robots in every industry. How does X YO's vision like fit into that and try to keep up with that?

Yeah. Yeah. A robot is the two things, right? It's it's a node which

3:00witnesses things, right? And then someone who does has something that does work and that work then you know someone gets paid for it. And so if you think about deepen right the satellite physical infrastructure network we spoke about you it on your show right where you have devices and for those devices if you put them up you know like you get money right you can do that for

3:25helium for example you create a cell network right or in our case you put up our devices you collect and validate data and you get xyo for that and

and [clears throat] there's many others you know there's a compute for AI and data storage and energy ones and many many different ones and so in a way you know robots the way Elon

3:48Musk envisions them they are just owned by Tesla right and then part of a centralized company you know but we are living in something called a deep decentralized physical AI basically where you are able to that robots are basically owned by by you and me and many other people Right? And you are able to participate in their in the value they create.

4:15And if you think about cars for example, right, in America, cars are used 3% of the time and it's a huge waste of resources, right? Well, for most of us, it's a it's a second largest purchase we make. And

and but if you your car would be self-driving and put go somewhere where it knows it how to charge itself right then suddenly you can use it 90% or 95%

4:41and it earns money in the time when you don't use it right and it becomes an asset which which earns you money it's the same thing with robots right if you are able right to send your robot out to I don't know law like or do garden work at your neighbor right and then you should get get just get the pay for it and I think that's going to happen and that's

5:06where we fit in right because we are we are connecting all these devices like proving that something is where it's really supposed to be right and we witnessing what's happening basically proof of work of the robot did the robot really do the garden work and then and then the you know the data gets written on chain right and then there are payment rails, right? So that your robot

5:31or you gets paid for this and you know all of this can happen in an automated fashion. It's very cool and I want to dive into this specific example more with autonomous cars as you know robo taxis start the numbers start growing and likely soon everyone that owns a Tesla will be able to turn on that function. Can you talk about the differences there

5:56and why we need or what are the main benefits of the DPIN network versus say I have a Tesla already, Elon's going to turn on the network. I'm able to make money. Maybe I don't have as much control, but I don't have to learn the technical details in the back end. What am I missing out on by not having my car in a decentralized version of that autonomous network?

6:18

Yeah, it's it's someone else earns off you in a way, right? this middleman we try to avoid in [clears throat] blockchain, right? We say everything you want to do is peer-to-peer with this hippie, right? Which believe, you know, there's a borderless world, right? And then we should all be able to interact in the way you want to without

6:38those intermediaries, you know, like intermediaries cost money and they're renting, right? They take value away from you. If you look at US healthcare, for example, right, of the premium, healthcare premium you pay, right? only 6% of that goes into care, right? 94% is wasted in the system. It's absolutely crazy, you know, and everybody's complaining why healthcare

6:58is so expensive. It's because it's it's not managed, right?

And imagine a blockchain peer-to-peer run insurance company. It's amazing, right? And the same thing if you think about your fleet of cars or you know, a team of robots, right? run in a decentralized manner, right? Where it's open source, right? You can plug your robot into the system and the

7:25platform, right? And so you can take it out of it and put it into another platform, right? And where does it work? And that's I think it's it's a fairer world and in this world also like every humans are are profit seeking, right? And so he's like you are the sucker, right? If you let the big corporations take some of your money,

7:52right? While you actually could earn more doing it yourself and that's it, right? It's it's control and being able to monetize. Better.

Yeah, it's a great point. Especially if you have that option. It's sort of like what I if you're looking at it in DeFi terms, having the year in finance of your autonomous vehicle. Hey, I can make more if I switch to this other

8:18network. I have the freedom. I'm not stuck inside of the Tesla network, for example. I can, you know, move to anywhere that I want because I own the ability to make money on my car and it's not just stuck under whatever Elon says,

right? Yeah.

And that example about the healthcare, I would love to see that as well. I wonder if you know

8:41because the percentage of expenses you mentioned a vehicle is sort of the second biggest expense. Health care is getting up there you know almost competing with that as well. And I know some people that are paying like tens of thousands when they have to insure their whole family. It's such a scam and it gives you a lot less value. Of course in emergencies. Yes. But at least your car

9:02you use it most days. Health insurance. Some people don't use it at all and they're still spending thousands.

Yeah. Yeah, it's absolutely crazy what's happening there. And I think it's it's a system where you know you have incumbents like protecting themselves with regulatory modes, right? And now it's very difficult for newcomers to get into there and to try to

9:26change the system because you know there it's crazy regulatory modes and that is because of decades of lobbying by the healthcare companies right to protect that turf and I think we see the same in crypto right now the discussions between banks and crypto companies right where where banks are saying no there shouldn't be yield on stable coins you know like arbitrary rules

9:49they shouldn't be this and they shouldn't be that, right? And it's because they try to make the rules which benefit them the most so they can make the most profit.

And that's a problem in these closed systems, you know, if it's banking or insurance or is it the robot fleets in the future if it's AI itself, right? And the metaverse, right, which

10:13everybody spoke about a few years back, you know, there was the open and the closed metaverse, right? the closed meta was the things suggested by meta for example right and then there were the open ones from sandbox and others and it's like I rather live in a world right where we can have choice to plug in or not to plug in and you know

10:36where not 90% of my value gets taken by someone else and I think many people will many other people will choose that as well

definitely but you know it's it's tough when there's a network effect of these centralized providers, whether they have control of the regulatory environment like the banks, you know, to throw it down to protect themselves.

10:57But even if they don't, trying to get adoption of the decentralized versions of these networks when there's already a network effect of something centralized is tough. H what are you know can we just compete on price making it cheaper or how else can the decentralized and obviously have more benefits but maybe more technically complex in the back end how can we grow those to a

11:21bigger piece of the pie

I think that's a critical question I think to be able to compete with centralized networks which are often quite efficient right or could be more efficient in some cases right like it's and The reason the way to do it is like I

think you have to make it very easy to onboard you know like the user interface like your grammar needs to be able to

11:47onboard right into that ecosystem and crypto fails often in that especially in deepin right you need to buy this device and install something on your computer and then need to connect it and something with your Wi-Fi and then you know you lost people and then you need to create a wallet you need to be worried that you don't lose your keys and then you lose people again

12:08and it's it's just not good. You know, it's the best way to do this is if you have, for example, a website, right, or an app, you know, download it or you log in and then and then everything happens automatically, right? Some preset settings and it runs on crypto rails or blockchain rails because that's more efficient in a lot of ways and decentralized, but that's in the

12:29background, right? the solution is first and on the on the cost side I think there's so much access capacity which we have like the car as I mentioned but if you think about your computer right you can use your computer to mine crypto today and use it excess capac capacity you know by doing for example data storage or data compute for AI training

12:53and but many people don't do it because it's it's complex you know but I think these networks you know they will grow over time and will be very very successful you know or if you think about access compute capacity in data centers it's same thing like ether made I think $140 million in revenue in the last 12 months right they had a decentralized compute player in

13:20deepin and it's amazing right because they take it away from like Google and Amazon AWS and all the others.

Definitely. Yeah, the AI is getting better. You know, I think our grandparents if they're still alive, would be able to use the basic AIs, but some of the AI agents, which I'm curious if that can tie in to the data data verification networks of XYO, they're

13:49still pretty complex. I recently saw the hype around the claude bot and I tried to use it myself and you know it required terminal and NodeJS and people are getting dedicated servers for it and I was like this is a lot of work. I don't think a non-technical person is going to be able to run AI agent with any you know it's like you need to

14:10have a degree in programming to figure this out. With AI agents, would would that technology tie into data data verification on XYO in a similar way that physical robotics would?

Yeah. Yeah. AI agents are are interesting in that they are complementing AIS, right? They are they are doing spe specialized functions, right? Or in a data collection and

14:35validation way. They might say, you know, an AI might get a feedback that it hallucinates, right? And then an agent might go and collect another piece of data and that agent in a in a in a DePIN network like ours can be AI as well as human you know it doesn't need to be

[clears throat]

so for example if there is no sensor there you for example

15:00aperture might be missing somewhere or does it rain right now somewhere right and so with our network right that request would come into the network and then you know someone would take a picture of the clouds and say no there are no clouds for example right it's sunshine it's not raining and then that data gets validated you know with proof of location so that it was taken

15:24at that location and something we call proof of origin in our case where it's you know we can prove where the data was generated on the phone for example then saved into an IPFS database and then put it into a smart contract and run through the XL1 XY layer one network and we are able to say where that originated and that obviously helps against the

15:51hallucination of AI.

Mhm. Yeah. Weather information is especially important right now as you know in January 2026 this crazy storm's rolling through the US and the news in the media we're we're hyping it up like you know it's like an Armageddon and we're going to lose power for a month and go raid the grocery stores. And people distrust whatever

16:16they said. And usually, you know, when they say stuff, they're o over exaggerating. And when they don't say anything, it's actually when the storm hit. And at least here in Texas, it seemed like they were really off the mark. Sure, it was it's cold, but like, you know, I don't I think that raiding the stores was unnecessary for, you know, a light dusting that just happened

16:39in one night. I feel like the original, you know, we've been speaking since 2018 location and data verification that ties in perfectly to weather as one of the earliest use cases for DPIN.

Yeah. Yeah. Yeah. you have a partner called weather exam and it's a it's a great use case I think because in lot of places you don't have the

17:08temperature sensors and other sensors you know which you need and to do good forecasting and forecasting is important for industries like farming for example you know and if you or electricity generation and these kinds of things and so in US obviously it's not a problem but if you think about you know some places in Southeast Asia or in Africa or on some

17:32remote islands right you don't have proper forecasting and without that you might make the wrong decisions on when you saw the seeds in the field right

because maybe you don't have electricity to pump water or your well went dry or you don't have a well to begin with and so

with deepin you're able to solve this in like you can put up temperature sensors

17:54Or you can you know you can take pictures of the sky and you know take pictures of the ground and if you have temperature temperature

feed all of that into a farming AI right it's it's all the farmers of a region you know let's say 3% only participate right and but you get a wider net right and then suddenly you get all this data feed it into farming

18:17AI and that gives you farming advice and you increase your yield and your probability that you don't have to go hungry next year and that's just super cool, right? That's deep in really changing the world.

It is very cool. And I feel like combining that data with prediction markets, something like like Poly Market, where there's money on the

18:38line, people are betting, you know, and that helps the insurance could tie into that as well. The percent chances that it's going to rain or that the crops will turn out right. I feel like it's sort of the perfect storm of like combining these technologies to get perfect information,

right? Yeah, I completely agree. That's a British markets are amazing. You

19:00know, it's of course smaller bets, you know, some people try to manipulate them, right? And there has been accusations of front running on some things you know if who predicted that Maduro you know Maduro is out of office some people made big bets you know who weren't visible on poly market before and they won stuff like that you know makes you sometimes wonder but overall

19:30it's it's so neat how the human psychology is profit driven as as we said before right so you try to maximize your outcome, your your profit outcome and so you don't make bets which you don't think will win, you know, and so it's the skew often in the right direction and gives the right results and it's quite marvelous.

19:56

Yeah, definitely. There's been as prediction markets are growing very quickly, I think maybe they need to flush out some of the some of the regulations and rules. it seems like, you know, insider trading doesn't matter there or something like that, you know, on who bets on what u with a little bit of manipulation from, you know, if this person says it, they read that

20:17before and they're like, I'm just going to say what the is against the bet to mess it up. So, that's something that maybe needs to be worked out. But in terms of having the data that people, you know, some people they're basing these bets on the prediction market off of data and whether it's trustless data that maybe is through the data verification network or not. So

20:40I feel like that's an important aspect to the XYO layer 1 network in having sort of trustless verifiable data in that it can be it's not prone to manipulation. Is that right?

Yeah. Yeah, you're right. So, it's like we can prove for example that, you know, Trump, President Trump was on a stage, you know, on a certain date, right? And then you know, there's 100 witnesses

21:08there, right? And that resolves the poly market bed, for example, right? And that is definitely something XO is able to do. we have some conversations about it. [laughter]

Yeah. Yeah. I'd love to see that. I feel like there, you know, as these markets get more liquid, it's becoming easier to trade and without, you know,

21:35losing money on slippage and things like that. You know, not everybody's a weather expert. And as these new event contracts come in more people should get more interest in it and the ability to make money and basing their decisions off of this trustless data. I want I want to go back to robotics for a minute because we talked about that a bit and it's it's sort of

21:58at the inception of you know eventually they're going to be in our in our house. With the information that the for example physical robots maybe in your house or on your property have and they base their decisions off of are there are there threats of like unverifiable data or hacking or you know where where could deepen fit in to make it more safe

22:25and secure and better.

Yeah, I think that it's definitely threats. You know, there was a there was a scandal a few years ago where Tesla was recording people in their garages and so on, right? And they the that data was verifiable. I think they had a Slack channel in the in the garage about like I think something about accidents or something, you know, like

22:52

and people shared shared that in the company and it was a laughing stock and that's just a problem if you have big corporations have visibility, you know, and then they decide what's private and what's not private, right? You lost your right to privacy, it feels like. And with blockchain you can build things like zero knowledge proofs right where

23:17you can pro where you can have the knowledge of some data is true incorrect but that you don't tie to an identity for example and deep works there in being able to have that node somewhere without you know that tying tying data to an identity and I think that's that's that's very very very important. And especially, you know, as you know,

23:45like robots might be in our bedrooms, you know, or they might wipe our butts. Who knows what, you know, [laughter]

this is next level from being in your garage. Now they're going to be putting you to bed and standing while you sleep.

Who knows, you know, like elder care, right? There's such a shortage of humans for elder care, right? I think

24:07if the wonderful nurses and helpers, you know, could get some help by robots, you know, then they might not feel as stretched and stressed, right? And they can focus on the task which matter like the emotional thing for example and so I think robots are important for that and but you want to make sure you have your privacy.

Mhm. Definitely. So say for example we

24:28had a physical robot as a caretaker in you know this centralized company maybe it's Boston Dynamics maybe it's Tesla versus a decentralized network I feel like it's especially for elders if it's for elder care you know to explain to them hey this is more trustworthy because it can't be hacked how how would you explain that to a grandma or a 5-year-old without

24:55getting technical in the reason that decentralized robot autonomous network and the data behind it is safer than the centralized competitor.

That's a good question I think because that's what we need to figure out. I think I would would say you know there's this example there was this many cams which were hacked also few years ago you know and suddenly there

25:18were all these pictures of babies and people having conversations and so on and it was embarrassing for the company but also quite damaging and you know if you don't want that to happen you use networks which are more protected and run on blockchain rails which is a I think technology which can protect your privacy.

Definitely.

Can can you elaborate on the need for

25:46specific XYO layer 1 for verifiable data versus just tapping into Ethereum or a chain that has a bigger network effect?

Yeah, our chain is specifically built for data. You know, Ethereum and others are made for value transfers, you know, where you transfer one E right to someone else, right? Any other cryptocurrency and ours in ours you can do that too but it's

26:15made for data specifically so it's not all about data storage and data analytics but the data certainty you know that you can prove that the data was generated on this robot for elder care right then you can make sure that there's zero knowledge proof so that you can't identify the identity right then you store the data in an unidentified on the layer one or on something else

26:40while the hash layer one and do something with that data maybe makes the robots more efficient because you learn something from the data or you know or do other things training an AI maybe to then build better robots and so it's it's the layer one creates immutable data right which can't be changed later right so someone can't

27:06say No, my the robots didn't do that work [laughter] and so you don't need to pay for the elder care or it's also you know to make sure you know that this no hallucinations and so on and connects it connects with the real world basically this reality engine I mentioned in the beginning is where it collects real world data and it puts it into this reality engine and then

27:33connects it with AI and other virtual worlds.

Yeah, proof of work is obviously important. If you're going to be paid for the work, you know, you need to have verifiable proof that it was done. But once the work is done, then you actually move to the payment. You know, stable coins are so important for on crypto rails, especially as autonomous

27:56vehicles and IoT devices are going to have wallets that are paying each other, whether the car is paying the charging system or you're paying the car. How does that would that work ideally on the XYO layer 1 network as well with traditional stable coins or do you have the option to use other chains while using the data from XYO?

say the question again I missed the

28:21beginning. So, if you're going to be paying with stable coins

or you know more volatile crypto assets as well, would that be ideally on tied into the X YO layer 1 network and do you have the infrastructure to pay, you know, with USDC and USDT conveniently or would you use other chains potentially and, you know, have the option of whatever chain

28:46works best? the verifiable data is on XYO layer 1, but maybe you want to pay with Ethereum layer 1 or something like that.

Yeah. No, you can do it with XY layer one in the future. The idea is that you can connect it to everything out there, right? it becomes this spideret which which connects and then payment rails for stable coin payments

29:09are are important part of that not only stable coins but also XL1 XY and other cryptocurrencies and in a way I think it's what we spoke about earlier it's just important that it's all seamless so right now you know like 20 years ago yeah you took out your your cash right if you needed to pay your friends pay for pizza right you need to get some find some cash right and give them the

29:32right amount they give you change. It was complicated, right? And today you have Venmo and casher and stuff like that, right? It makes it more convenient. You can send the immediate the complete amount and it's in a virtual form and you don't need change, but it's still cumbersome in a way because you need to figure out how much you owe them and you need to take

29:55out the app and you need to know who to send it to.

in the future, right, everything will be automated, right? where is these robots and connected machines will be saying like they will send an invoice basically to your system or wallet right and then you know there is x yo proof that pay that work really happened right and then that

30:20payment will be automatically sent to the machine or to the owner of the machine and it's all going to happen automatically you don't even need to worry about it you know it could even go as far that in the smart city, right? You have parking spots, let's say with dynamic parking fees, right? So, whoever pays the most, for example, has the right to

30:41that spot, right? And then and so you have you just define what the limit is or you negotiate right ofways in wherever, right? On streets, right? And cars negotiate the right way just themselves. And you say, "Okay, I have 10 more minutes, right? And I'm fine to go slower and I make 10 bucks on the way to work because I let all

31:04these other guys go first, right? And that's a good way, right? To make the world more efficient for yourself. Yeah,

definitely. as we, you know, start off 2026 with, I think a little bit of optimism on the crypto side, and a lot of optimism on the AI side. You know, we're

31:32seeing the AI is continuing to grow. What are some of your predictions for how these two industries will continue to merge and what innovations might we see and growth towards actual adoption of these kind of verifiable data networks outside of the echo chamber of blockchain? Yeah, I think AI is just hungry for three things, right? They need storage,

31:57data storage, compute, and data, right? And they get that in any way they can get it, right? It's it's such in the US there 1% of GDP is already generated by all the AI stuff, you know, by those three things basically. And so you know and if it's decentralized or centralized right now those systems don't care right they just need it and so I think blockchain can just be

32:28an awesome part in serving AI in that way and then in the future I think you know blockchain and our layer one and then and other technologies but can make sure that the AI is doing what it's supposed to do and doesn't hallucinate and you can prove that it's doing the good thing you know and I think if you look a few years out right a lot is coming you know like

32:56quantum computing for example you know they say the first commercial machines coming in 2027 and with that you know we're going to see an acceleration in AI and its capabilities and with that you know we need even more data and we need everything is going to grow right and with that you know we going to better robots, right? And robots going to go on

33:17to going to be space fairing robots, right? And they're going to do all the healthcare things, right? We solve diseases maybe and all all all the things we hope for to solve, right? All the problems of the world basically. And for that, you still need, you know, compute, right? You need data storage and you need data. And so and blockchain fits very nicely into that. I agree and

33:46I don't think the AI cares you know whe whether the AI whether the information is decentralized or centralized from the source but from what I understand with these deep pin networks you can gather so much more personal localized locationbased data with the deep pin networks and everyone on their edge devices and the 10 million nodes already on XYO and if there's an incentive there

34:11through the cryptoeconomics for them to contribute information, the AI is going to suck that up eventually. Because there's there's just a lot of information in the world that is yet to be documented and verified. So I definitely see the role that blockchain and deepen has to play in that.

Agreed.

That sounds great. Well, I'm looking forward to seeing the next major

34:36partnerships and updates. I know that XYO has a lot in the works. I appreciate your insights into AI and robotics. We didn't even touch on space but I'm sure locationbased in space maybe for next episode we should touch on that because you know with the SpaceX IPO a lot of people are interested in that and there's a lot of location data in space as well. So let's keep

34:57that one in the back pocket. All the best to you and the team and all the people that are continuing to grow XYO layer 1. You know, we can mention there's lots of easy ways for people to start contributing data or using any of the applications. I can leave a link in the show notes below for if people want to test out some of these apps that allow

35:22them to earn XYO tokens and contribute to the verifiable data on the networks and check out XL1. I'll leave all those links in the show notes. Thank you so much Marcus for coming on and would love to follow up again in the near future.

For sure. This was fun. I want to talk about space with you on the next one. [laughter]