XYO Layer One | Quantum AI Wallets

[00:00:00]I've been trading crypto, crypto, crypto. I've been trying to get more every time he dips low. I've been trading crypto, crypto, crypto. I've been trying to get more every welcome to the Chain League podcast. I'm your host Joshua Roomsberg. Today's episode is about XYO layer 1 quantum AI wallets. Today's guest Arri Troll.

[00:00:27]Welcome to the show, brother.

[00:00:29]Thank you for having me, Josh.

[00:00:31]>> All right, so let's dive into what X YO has recently released with the Quantum Ready AI wallets.

[00:00:44]What does that even mean?

[00:00:46]>> Well, so we added um quantum support for the new algorithms um based on the NIST suggestions um for quantum signing basically for our XL1 chain. We haven't released a new version of the XL one chain yet to use that, but you can sign um datagramgrams and that sort of thing with the wallet. So what we did was we took the NIST standards that they came out with fairly recently and we added added um MLDDSA 65 uh signing to our wallet. It does have some costs to it.

[00:01:15]So there is some considerations for when we want to have it actually starting to use it.

[00:01:21]>> Okay. Now with that, are you guys building a standalone wallet and is it exclusively for the XL1 or will it be supporting other blockchains?

[00:01:35]>> Um, we've considered supporting other blockchains in the future, but they have to declare what their um their settings are for their postquantum um addressing and stuff like that because we were u originally using very similar signing to Ethereum for our signing.

[00:01:50]>> Okay. you know the ECC elliptical curve signing there and u we wanted to get prepared for people to to change over to using quantum resistance uh signing and our our algorithms have been or our protocol has been designed from the beginning to be able to support multiple signing algorithms with this in mind. So a person can choose you how much they want to secure a c certain thing whether it's a a datagramgram whether it's transferring of money depending on the importance of it uh because there's a cost to using quantum resistant signing you know in size especially because the signatures are substantially larger. So it's not necessarily a person wants to use it for every single one of those but if you have a large amount of of tokens for example in a wallet you may want to use uh quantum resistant there. Um, and you don't we won't probably won't know when it's required until somebody has cracked some encryption and then it's probably going to be too late. So, people probably should start switching over to using quantum resistance wallets sooner rather than later.

[00:02:47]>> Okay. Now, will you guys be launching an Android application or is it a website UI?

[00:02:56]So currently uh we have uh um our SDK has it supported in there and we're um launching our um our Chrome plugin which is kind of like Metamas which is a wallet for that which will have the signing in there.

[00:03:11]>> We're also in the process of launching uh Android and iOS um native application which are wallets as well which will will support this as well for XL1 and so you could use it from any of those.

[00:03:22]Okay? Because me personally, I spend 90% of my time on mobile. So, for me as an enduser, if I'm ever going to use anything quantum resistant, I want it to be able to be accessible on mobile. Um, one thing I want to point out here is just in these first couple minutes, you've already mentioned there is an increased cost to using this quantum resistant technology. Let's go into that a little bit and what this cost entails and how do you incur higher costs with these transactions.

[00:04:01]>> So, the the basic difference is the size of the signature. You end up with a much larger signature. So for example with our current um you know ECC um or lift of the curve um signing the the signature is about 32 bytes and in the um the NIS suggested you know it's the um the NIST um was it the there's no number for it um I can't remember but there package the the FIPS 204 the FIPS 204 from NIST >> yeah 65 in it um from from that that um the MLDDSA65 which is actually one of the the smaller ones the signature ends up being about 4K. So you go from 32 bytes to about 4K which is a substantial jump.

[00:04:48]Um >> that's nast.

[00:04:50]>> So I think it's one of the reasons why people like Ethereum are kind of dragging their feet a little bit because if every signature that goes into their blockchain is you know it's it's more than one or >> size. Yeah. That's going to be >> substantially bigger. So all their, you know, their their blocks are going to be bigger, which means and since they're in a situation where they have to store every single thing that um that's on there, it's going to really bloat their their space a lot. And so >> Oh, and them Ethereum transactions would be crazy. If we hit a hundred without it being a 4,000, that'd be crazy. You're talking a three $400 Ethereum transaction at a minimal >> when you start talking fast. Yeah.

[00:05:30]>> Yeah. That's nuts.

[00:05:32]>> So, do you want to start every transaction with with, you know, three or 4K of of bytes? Probably not. But, you know, if you have a big wallet, then you you probably do want to do that. So, it's >> to me selective signing and allowing a person to sign insecurely for non-important things makes sense because, you know, no one's going to spit up a bunch of quantum computers to go and crack, you know, my one thing which signs the temperature at my house today. Right.

[00:05:58]>> Right. Yeah.

[00:06:00]Okay. So for the big infrastructure on quantum resistance, what types of customers or developers would you like to introduce this technology to? Because like you were just mentioning, regular people aren't necessarily going to sign every transaction with a quantum resistance to it. So, what types of what types of customers would you like to acquire?

[00:06:34]>> Well, what I think um everyone's going to need this um not just the big customers out there, but like take for example a home user who has um say you have, you know, $1,000 worth of of XL1 in a wallet, you may want to assign that with a quantum resistant signature and use a quantum resistant wallet for that basically because the way it works in our our case is you make a different set of set of addresses basically for quantum resistance. They look a little bit different. They you know they look a little more like Bitcoin's addresses. So you make that address, you transfer your your funds to the quantum resistant one and then um what you would do is you would make smaller addresses that have a smaller amount on them. So you can use your traditional signing on there. So on a day-to-day basis, like say for example, you're using, you know, something like OpenClaw and having OpenClaw go and, you know, do things on your behalf. You wouldn't give it your entire treasury to go and do it with.

[00:07:25]You say, "Well, I'm gonna send, you know, $5 worth of XL1 to this account so he can go and pay for stuff." So it's not a honeypot at that point. You know, no one's going to go and try and crack that. Um, you know, where like Satoshi's wallet on Bitcoin, >> it's a pretty big target, right? Like, you know, if you can crack that guy with a bunch of computers, it's really worth your while. So >> yeah, >> I think what most people going to be doing is it's kind of like like when do you make a um a cold wallet and put your money in that, right?

[00:07:48]>> So it becomes like a tiered system. You you want to have a cold wallet potentially uh for the important things, but a cold wallet that's not quantum secure is not going to be a great cold wallet in the future. So my guess is in the beginning most of the cold wallets people are going to have are going to be quantum resistant ones where you sign every once in a while with it because you want to take money out of your treasury and then after that you basically go and use the the less um intense um signing u systems and so I would think most people would want to use it once in a while. I can see larger institutions like you know exchanges and that sort of thing um using it much more often because they just deal with much larger amounts of money you know would care about that because you know getting into a bridges treasury very valuable for hackers. So, so if you're if you're a target basically that's when you want to use it uh for the most part in the beginning eventually it'll probably become um most transactions because as you know storage gets bigger and um you know as as basically three or 4K becomes less and less of a problem for people then why not just sign everything with it right >> because at some point 20 years ago you know a 32 byt signature would have been prohibitive for people. Yeah, for sure.

[00:08:56]I seen something the other day. I was looking at uh old cell phones from like 15 years ago and the size of memory versus what we have today is mind-blowing. Like uh two videos and you'd have all the storage used up on an old cell phone. So that's the same way you have to look at blockchains now. the amount of space inside a transaction. As time progresses, as these networks grow, the amount of memory that can get stored in a hash increases as well as these networks grow. Now, another thing with the quantum resistance, you had a good point talking about the bridges. Do you feel personally bridges could minimize the risk for hacks by using something like this quantum resistance signing?

[00:09:56]Definitely. Um what we do for our bridge um is we have different tiers of bridging. So we have like a um a we call a slow bridge which basically you can lose can move a large amount of money with um but it takes a consensus vote for example for that money to move. So it takes about a week to move the the money in that bridge. And then we have our fast bridge which can you know bridge money for you from XL1 to Ethereum you a wrapped XL1 on Ethereum um and back in a matter of seconds in a normal transaction but that one's pool that it's it's dealing with is much smaller and so I think the first step for most bridges is to have different tiers and you know different speeds of of bridging. And we use, you know, consensus versus um just plain signing for our two different things right now.

[00:10:42]But I can see definitely a three- tier system in the future where we have the the super slow consensus model and then we have a a medium one which is basically, you know, a little bit more expensive, but it's going to be for larger amounts and maybe, you know, you put put a threshold in your in your bridge whereas if it's over a certain amount or if your pool is is larger than a certain amount, you use your wallet um in in quantum mode basically. And if it's if it's small and insificant, then you can basically use it um in a lesser mode there. But that's definitely a case, you know, whether it's a treasury for an exchange or for um a bridge, you know, it's definitely going to be important for them to to you know, not not get get hacked by a quantum computer.

[00:11:24]>> Now, let's circle back to the quantum resistant XL1 wallet. Now, when you create a wallet, do you create a 12 or a 24word seed phrase? Or have you guys thought about pass keys like Google and some of these others big corporations have already started to integrate pass keys that are saved directly on your device?

[00:11:54]>> Yeah, we definitely have thought about that. We're we're actually in the process of introducing that in our um our wallet on on Chrome. The way that actually works, at least in our case, is we do the 12 or 24word keyphrase and we do the same thing for our quantum resistant wallet, but then you can save it in a a bastion, basically a secure bastion, which is like the keychain, for example, on a Mac or um Last Pass or whatever bastion you want to use for it, and then you unlock that bastion by actually using your fingerprint or a key a pass, right?

[00:12:27]>> Yeah. And so um really it becomes a two-step thing there where instead of like right now you log into MetaMask with your password. You can just unlock your wallet then with the um you know a pass key instead which is way to go. But at the end of the day we have to get that um that the 12 or 24word uh phrase because that's the thing which you know when you hash it you basically get the the addresses which you're going to use in the private keys. So um you know that it's unavoidable to actually have that.

[00:12:55]It's just it makes it easier to not have to remember it, right?

[00:12:58]>> Yeah, that's exactly that's exactly what I was thinking about is when you think about quantum resistance and what that brings to the table, I'm trying to think on the big scale of things. What does this bring that we haven't already experienced? You just mentioned that we still have the 12 or 24word seed phrase.

[00:13:21]Is there anything different in creating a wallet that is quantum resistant versus something we've already seen in the past?

[00:13:30]>> Well, so the the user experience is actually very similar. Um, you do the same thing with that keyphrase. You you get the keyphrase, you can store it, you know, in a um a last pass or something like that if you want to. Um the the size of the address actually is pretty much the same. um because you know even the addresses we use in Ethereum or on XL1 right now they're shortened versions of the the public key and so we do the same thing for this you know so those are not things which are really at risk um nearly as much for quantum uh problems >> existence yeah >> as signing really u because I can guess your guessing your public key is not that important because it's well the public key the thing which you know and also hashing for example is not that much of a difference like for example you know instead of using Shaw 256 we might use Shaw 512 to hash things which would double the length of that but doing the SH 512 is a a substantial improvement over SH 256 in in hash size.

[00:14:22]So hashing is not particularly at risk for the quantum um attacks. Quantum attacks specifically go after the ability to sign on behalf of somebody else without their permission. Right?

[00:14:35]And so if I can know what your public key is, I can go and u make a signature on something that looks like you signed it and you didn't actually sign it or you you kind of did actually sign it because I actually got your private key what it is, right?

[00:14:49]>> Yeah.

[00:14:50]>> And so that's what we're really defending against and that's why the signatures get longer there. But we try to leave the rest of the experience the same for people. The biggest difference is you have to when you create the account, you have to pick, do I want to have a wallet that has accounts that are quantum resistant or ones that are legacy? And um then you can basically move them from back and forth from there. There's not a system that I know of that we could use where we can just start signing with quantum resistance signatures on our current accounts because that's what I think people would have loved to have where it's like, well, don't change anything is just push the quantum resistant button on your wallet and voila, everything's there.

[00:15:25]But the problem is that it's not really the signature that we're defending. It's actually the methodology of how you sign with your private key and how big your private key is. And so if we don't change your private key, then there's no way for us to actually protect it. So we have to change your private key with into a different format >> to make it work, which makes the address change. And so that's why you can't really like retrofit a legacy addressing system to make it quantum resistant.

[00:15:53]What you have to do is you have to basically just make new wallets that are quantum resistant and then transfer your stuff there which is actually an interesting problem for things like Ethereum or Bitcoin. So if you go >> Yeah, you just blew my mind when you just said that. My mind started racing thinking about all the Bitcoin developers that always said well when Quantum comes we could just implement that onchain and everything will be copetic and you just said no that's that's not the case. Well, it is copathetic going forward if everyone starts using those addresses, right? But now, like Satoshi's wallet, you can't call up Satoshi and say, "Hey, by the way, make a quantum address and trans transfer your money over to it, right?

[00:16:33]Because that's what you'd have to do."

[00:16:34]So, like I don't know how they're going to treat that as far as are they going to lock down those addresses and say, "Well, >> you know, if someone's trying to move money from Satoshi's wallet, then you know, our producers will not allow for that." Or are they going to treat it just as like a a treasure hunt?

[00:16:47]>> Finders keepers. Yeah. Look, you know, if you can crack Satoshi's wallet, it's yours. And so, it might be that there's going to be, you know, once the quantum world comes around, there's going to be this great treasure hunt and anybody who hasn't moved their money because there's plenty of addresses, especially on Bitcoin, >> that are are lost or in landfill somewhere or yeah, that have tons and tons of value in them and they're very valuable for a quantum person to go after. And if those people can't can't upgrade their their wallet basically by moving their money from the old wallet to the new wallet, um it's you know it's going to be fair game is my guess.

[00:17:22]>> Um I haven't talked to anybody from the Bitcoin Foundation or anybody like that who hasn't, you know, said that like what's their policy going to be on on old dead wallets, but that's where I think it's going to be an interesting thing which you know could be an interesting game basically. It's like well you know it is what it is and you know have at it.

[00:17:40]>> Yeah. No, that's mind-blowing because I think about that all the time. Even just in this last month, someone in New York was they filed a lawsuit to recover I some crazy number of dead wallets on the Bitcoin chain >> and it was wallets that haven't moved and I want to say it was like thousands of wallets that this person put in a lawsuit against trying to take ownership of these wallets. And the joke online was, yeah, even if he claimed ownership, he still need the private key. So he he even if he wins the case, he only wins half of the battle because you still need to get into the wallet even if you own it. So you still need the private keys to get into each of them. But then when quantum comes, if that guy does win this lawsuit and owns them dead wallets and then somehow he comes up on the quantum technology or teams up with somebody that has it to crack those wallets, I mean, that's a whole bees nest you're coming into over the next 5 to 10 years whenever that happens.

[00:18:46]>> Well, that's probably what what his goal is. That's actually pretty genius of him to say, "Well, I want to claim ownership of these because if everyone can can crack it with a quantum computer eventually, I legally have the right to to to take those or whatever it is because I can only see if a person were to take Satoshi's wallet with a quantum computer, the first question is going to be, well, if you're not Satoshi, this is theft. If you are Satoshi, it's not theft, right?" And so now you get prosecuted for theft or for or whatever it is. But he's basically trying to set the table where if he does actually correct these things with quantum technology that it's not theft under US law or wherever he's at. And so it's pretty genius of him to go and try and preempt that that discussion of well who actually owns this. U I doubt that you know if his if his plan is to go and ask this like the um the producers or or the miners of Bitcoin to hard fork so they give him you know they transfer that that money to his account. I doubt they're going to do that. I just I can't I can't imagine there's any chance that they do that unless they're somehow really forced legally to do it. But um I don't think anyone's ever successfully >> used the courts to force a blockchain to transfer money without a key to somebody.

[00:19:54]>> Yeah. Yeah. No, that'd be nuts. So to transition into, why is your focus on quantum technology and the quantum resistance at this very moment?

[00:20:10]A lot of it to me at least personally is, you know, when when will a day come when quantum computers can actually crack addresses? Like >> yeah, >> is it here now? Like is there enough, you know, like if you look at at what what um Musk built with his uh giant data center with I think it was 200,000 H200s that are um are there. The amount of money that's available to build things like that is and you can build that really fast, right? So if the technology is there to to get quantum cards on oneu servers and you can load them all up onto a a a giant data center and network them together. I wouldn't be surprised if today if somebody had enough money that they wanted to go and set up a quantum network of some sort to be able to actually correct it. But I'm not in that industry deep enough to know. So the way we're going to find out you know the man on the street happen is when it happens, right? So, you know, like when's the the news article about Satoshi's wallet getting moved going to happen? Is it is it next week? Is it next month? Is it next year? Is it next decade?

[00:21:16]>> I don't know. And since I don't know, I'd rather prepare now because, you know, if I knew there's like a deadline, if someone said to me, you know, I I can guarantee you you have four more years.

[00:21:26]Well, fine. Let's wait two more years, you know, three more years and do it. I just don't know. And you know, unless somebody can tell me exactly when when, you know, the quantum day comes, you know, it's there's not that much cost in actually, you know, being preemptive about it, especially if you're only going to move your your high value assets into those wallets. That's why I think it's important also to not deprecate the the legacy wallets very quickly because um not everybody wants to necessarily go that pace. So that this way each person has their own choice. If you think you have like 10 more years, go ahead, use the legacy wallets, you know, that's up to you. The one thing which we do have to decide though is um on the network u we have to switch our producer signatures from legacy to quantum at some point and that's what's going to um happen probably before most people start using the actual wallet themselves because um we don't want necessarily for a person to be able to rewrite the chain either which is a different issue than um than stealing from wallets.

[00:22:23]if chains it's a little bit different for Bitcoin because they use um the hash which is basically um made um by their hash rates and stuff like that. So how they actually have their their linkage to the previous one is not signature based where like Ethereums or ours is signature based and so um I think it's a non non um a non-p proofof work system you have to worry about the signatures of the chain itself being broken. So I think um XL1 moving to a crypto or sorry um a quantum resistant basically system for the the producers signing the blocks um is going to have to happen probably fairly soon. We'll probably do it in the next six months or so just because that is also a target for people to be able to rewrite history. Um and I imagine Ethereum is probably you know pondering the same question of like you know at what point in time do they you know take that jump because it's it's as I says before it's not painless. So, you know, you don't you don't want to do it so soon that basically it just drives up your cost for no reason, but you don't want to do it so late where it breaks the system. So, when's the right time?

[00:23:28]It's hard to say. When you talk about block producers moving to the new quantum resistance signatures, would they still be able to process legacy transactions or is everything from that point forward have to be on the quantum resistance signatures?

[00:23:53]Well, for a period of time at least, probably for quite a while, we would still allow the processing of um >> both >> of both, right? And and you know, at the same time, if a person wants to use the legacy signature, it's again, it's it's their account. It's it's their choice if they want to do that. Um we've actually thought about for especially for IoT edge systems going the opposite direction, making something which is even less secure because it's nice and small. So, you know, we can do something that's, you know, a tiny little little address with a tiny little signature because all you want to do is basically roll these things up and then every, you know, thousand blocks of your of your side chain go and sign this with a a better signature and then finalize it and move it up.

[00:24:35]>> So, there are arguments for memory constrained systems like a a small Bluetooth device for example, using a a small algorithm that's not CPU intensive and is is space conserving. Yes, it could get cracked, but for someone to go and, you know, change the the readings on my water meter, like there's not that much of a risk of that. So, I think the the flexibility of use the the right amount of of security for your transactions is really what we want.

[00:25:04]Now, I don't think we'll allow people to put transactions on XL1's main chain with less than our current legacy um wallet size or wallet type. Um but for side chains definitely would make sense.

[00:25:17]But um yeah, I don't I don't think it's it's probably, you know, at least three, four, five, maybe even 10 years from us saying no, we absolutely won't run um legacy >> legacy transactions.

[00:25:30]>> Okay. And now another thing is I want to say it was two years ago I saw a blockchain I was working with or it might have been just a regular project but they they implemented Ethereum signatures to lock and unlock their door a deadbolt. M >> so if you could take that concept and you think about home security, I think the quantum resistance would be very important once we get to that level of blockchain tying into your home security. If I was signing a blockchain transaction to lock or unlock my door, I just open up my phone and I click a button and it signs a transaction whether it's on XL1 or it's on Ethereum.

[00:26:29]Quantum resistance would be very important. Say you have valuables in your house and if they have a quantum computer, they can crack into it.

[00:26:38]They're in your house and they don't need to break a door or a window. So that could be crazy.

[00:26:44]>> Well, the one thing which helps us there a little bit is that the odds of a a thief having a portable quantum computer that in real time can go and crack your your lock there. That's pretty far off still because um you know when we're talking about cracking Satoshi's wallet, you're talking about a very expensive large installation of a quantum system.

[00:27:01]So um you know you would have to >> you'd probably need a warehouse or two to put that together. Right.

[00:27:06]>> Right. So you if you're breaking into Fort Knox and Fort Knox has a locks that has you know yes definitely it's going to be a problem but if you're breaking into most people's houses like the you know yes to you or me it might be valuable but to compared to so that's kind of the whole basis of blockchain to a large degree is you make the cost of stealing something substantially higher than the thing's worth right >> so like if I wanted to I could actually I can rewrite the history of of of Bitcoin but what what does that require it requires me to have more hash rate than the rest of the world. So the cost of actually getting more hash rate than the rest of the world is so outrageous that >> whatever I'm stealing is not worth it.

[00:27:46]It doesn't make sense, right? And so as long as that that cost to benefit balance is what it is, it you know that's the the important factor for which type of encryption you want to use, which you know, which type of signing you want to use. And so as long as you keep inexpensive things, you know, inexpensive lock boxes basically, then no one cares. Uh, right.

[00:28:07]>> When Okay, let let's just take a hypothetical on that. If somebody wanted to put together the funds to crack a quantum resistant infrastructure, I mean, are we talking millions of dollars, hundreds of millions? what what what type of infrastructure would it take in financing behind just to give the end users an understanding of how secure their wallets would be.

[00:28:40]>> My understanding of the current um NIST systems that they're predicting right there is basically all the computers in the world currently combined cannot crack. It's kind of it's the that's kind of where we were with Bitcoin, you know, 15 20 years ago, right? It's like like if you take all the the computers in the world combined. Literally, you have all the money and you can buy all the computers and all the energy to run them and everything, you still don't have enough to actually crack it, right? It's like like more than the grains of sand on the on the planet and more than the stars in the in the sky sort of approach.

[00:29:10]>> No. Okay.

[00:29:11]>> So, but that's today, right? But at the same time, I did the math on my my one Dell server that's was built in 2019.

[00:29:18]It's it's seven years old now. And that thing is a million times as fast as my first computer, my TR80 model one, it's actually more than a million times as fast. It's like several million times as fast and as much storage and CPU and everything.

[00:29:30]>> Yeah.

[00:29:31]>> So, you know, over what is it uh 30 or 40 years of basically um of time >> technology advancing?

[00:29:38]>> It's it's gotten there. So definitely in 40 years from now, you know, if if there's a quantum computer that's several millions of times faster than the current quantum computer we have, then we get to the point where there's, you know, some subset of the of all the power on on the planet can actually crack our quantum things. But I think that's one of the things that that they're doing is they're looking at that and saying, okay, well, you know, we have this, you know, FIPS 204, we have FIPS 205, which is a little better. So So we have a path to it's kind of like we did with the old technologies. We'll just add more bits, right? They say well you know at one point like you know Shaw 64 was was fine and okay well we can crack Shaw 64 Shaw 128 and if you look at SSL like you know SSL lengths have gotten longer and longer over time because computers have gotten faster and faster and bigger and bigger and we can we can crack those we just added more bits and my understanding is that there there's a fairly good path for the quantum development where you know if we do that you know we can do the same thing there but now we're going to go from a three or 4k signature to a 40k signature eventually. 40 50. Yeah.

[00:30:38]Right.

[00:30:39]>> So, as long as you're willing to to pay enough bytes for it, it'll still be quantum secure, but there's cost.

[00:30:47]>> So, for an end user to understand this more cleanly, as the technology advances, the security and safety behind it is going to advance as well. So you wouldn't still be stuck on the first iteration of quantum resistance. You might be at level 50. By the time level one is crackable. Then at level 50, you have to wait until it gets to level 100 before that level 50 is crackable.

[00:31:19]>> Yeah, that's that's definitely true.

[00:31:20]That's one of the reasons why one of the things we did in our new addresses is we have a prefix in our address now. So our previous address is there was just a string of of hex characters basically.

[00:31:29]>> Okay. So it effectively has a null prefix but now we have a prefix on our addresses. So we say okay well if there's no prefix it's the legacy one.

[00:31:36]If there is a prefix it's a non-leacy one and that prefix tells you well which algorithms we're using. So what'll happen is when the next version comes out we'll have a different prefix for that and so on so forth. And so you know as time goes along it's kind of like SSL. have SSL version one, version two, version three, you know, it's like and so we'll have to, you know, keep on going with that and then, you know, people have to opt into those new addresses as go as time goes along. And so um like any technology, it's you know, once it gets, >> you know, decades old, it becomes almost um you know, so weak that that it's it's undefendable and it's always going to be a moving target. Same thing with AI, for example. you know, we have AI security problems now where people like, well, >> yeah, >> yes, but we have to then build AIs that defend against that or we have to and so these moving targets are getting faster and faster, especially with AI that um you know, it's going to be an interesting time the next few decades as far as the chase goes where, you know, can the the the the white hat hackers, you know, outpace the black hat hackers basically, whether it comes to quantum resistance, whether it comes to AI security, you know, all those sorts of things. It's it's it's definitely very very very fast-paced. Um >> yeah, >> nowadays.

[00:32:49]>> Okay. Now, to circle back on the wallets for these new XL1 quantum resistance wallets.

[00:33:00]You mentioned that they would have to send their tokens to a new wallet address that would be created. Do they need to bridge into the quantum resistance framework or is it the same token you just send it to a new address?

[00:33:16]>> It's the same token. So what'll happen is on our chain where the we have the address field right now for who you know who made the transaction or who signed it and that sort of a thing.

[00:33:25]>> It'll just support the the new address type as well. So if it has that prefix and it has the quantum resistant address, that address will receive it.

[00:33:32]It's the same as a transfer. Like if I transfer from wallet A to wallet B with the legacy ones, this will be basically transferring us from wallet A to wallet B, but wallet B is a quantum resistant one and the wallet A is a legacy one. So it's just a transfer inside of our system. Um now a separate question would be what are we going to do with like the ERC20 wrapped ones that are on Ethereum?

[00:33:52]Because uh we have to wait for Ethereum to figure out well what how are they going to deal with that on their system and then we'll have to potentially make >> uh ERC20. that's that's quantum resistant on there or or support that.

[00:34:05]So, we're only talking about our own chain where you do native XL1 to native XL1 transfers. Uh we can definitely deal with it there and make it really simple.

[00:34:14]There might be some bridging involved once it gets to other chains.

[00:34:18]>> Okay. Yeah, that's what I was thinking about is like um moving XL1 between Ethereum to the XL1 chain.

[00:34:30]you would have to bridge back and forth there. But on XL1 itself, the wallets are interchangeable.

[00:34:38]Correct.

[00:34:38]>> Correct. They're just they're just basically is a transfer. There's a new type of wallet that you know works just as if it's old type of wallet. So um it's almost like uh you know adding you're going from IP4 to IP6 where it's on your computer. Your computer understands both of them but then the routers have to support them and that sort of a thing going out.

[00:34:57]>> Okay. Now to move forward in the interview, let's go into X YO AISDK and talk about that a little bit and what that entails.

[00:35:09]>> So one of the things we found is um with AI coming about now, a lot more people have access to being able to develop fairly complex software and people want to be able to do their their own thing and we saw this on Ethereum for example.

[00:35:24]A lot of people struggled with solidity and say well because they were kind here's solidity do whatever you want and it's like some people did it's really cool but then the things which really shined was like in ERC20 for example where it's like oh just go and you know put in three values here which are you know your symbol your name and the supply for example and you know who who gets the first um supply and you can have your own 20. So it's more of a configuration sort of a thing. We wanted something kind of in between where it's like, well, we want to have more customization than just configurate configuring something. Um, but we also wanted to make it as as simple as possible. And so what we did was we we made a SDK that has uh skills for a variety of different uh parts of our system. So there's knowledge about XL1 and XYO. There's knowledge about how to use the SDKs for XL1 XYO. Um the JavaScript SDKs. At this point, we're going to be adding the um iOS and or the Swift and the Coslin um SDKs as knowledge bases on there as well. And if you give all these different skills, there's also like some patterns of how to do certain things and um how to use like for example we have a concept where it's the move move um redeem concept. So basically you can have I post a private payload, you post a private payload and we um then say okay well we've both locked in and then we redeem to see who won. Right? So, it's kind of like I play a card and you play a card face down and then we both flip them over at the same time. So, that way, you know, you can see what happened, but um but you can't undo it. You can't lie about what's there. And so, some paradigms like that we have in our our AI SDK as well. So, we can say, well, if you wanted to do a a move, move redeem or move, move, reveal system for your game, this is how you would do it using XYO or XL1 as opposed to having a person have to try to figure that out themselves. And we've really um seen great results with that.

[00:37:12]Um, some developers here at the company have, >> you know, in a >> matter of a week made, you know, six to eight functional games with our AI SDK really fast that that work and can, you know, work and and have a little website that can, you know, show people what's going on. It goes on the blockchain.

[00:37:29]>> It really accelerates those things. And so, I think the ability for adoption for people uh with the AI SDK is much higher than just giving them a a a traditional SDK like a JavaScript SDK and say, "Here's documentation." You know, first thing you do is spend six hours reading the documentation, >> trying to figure it out. Right.

[00:37:46]>> Right.

[00:37:47]>> Okay. So, what types of applications or products would you like to see or do you expect people to build using the AISDK?

[00:38:01]>> Well, there are some um you know traditional ones where you just want to put data on a blockchain for example. So if you know I'm a a company it's more of an enterprise thing where I want to have an audit trail for information coming from trucks that I have or that sort of a thing or sensors that I have in different places and I want to be able to roll those up and put them on on chain. Those are going to be things which I think we're going to see quite a bit of where it's I want to integrate this into an existing project as a way for me to be able to secure my data. Um I think for consumers or you know I say call weekend warriors the people who want to just uh you know use interesting things and do interesting things. I think the first things that people often do are um things like you know games or NFTs you know making different NFTs um because we also have an NFT and a and a um ERC20 type system uh we use a ordinal system more like like Bitcoin than the system on Ethereum but letting people you know make you know customize those things as well. I can see them doing that. And so, um, we want to have a showcase of our own things we've made and so they can see the different games, different ideas we've done. And then we want them to come up with their own ideas and and, um, you know, have them become reality in a matter of hours. Um, the the one of my biggest concerns though is at some point if Claude and Codex starts cutting people off because their cost to actually supply the AI, which we're using right now to do this, might be too high. they've already kind of fired a warning shot with uh Fable 5 where well it's free for two weeks and after they're gonna have to use tokens to actually pay for it and so if a person has to pay $1,000 you know to make a custom app on Xio because that's what the token cost is going to be from claw from um from codeex that's going to be problematic so we're going to have to explore u getting our AI SDK to work really well on llamas that are self-hostable and u you can do like in open source different um idees u without having to go that deep, which I we're pretty close to as well. U it still takes some hardware for a person to do it. You know, you're not going to take a 10-y old laptop and do it or anything like that.

[00:40:06]>> Yeah.

[00:40:07]>> But if you have the ability to host a quantized um model that is 80% as good as Claude, for example, you will still be able to use our our SDK.

[00:40:16]>> Okay. Yeah. Because I'm thinking about um specifically with Fable, they just cut that out. Now, is there any vulnerabilities you could see with an AI model like that against not even just XL1, Ethereum, Cardano, XRP? Are any of these ledgers vulnerable against a product like Fable?

[00:40:50]>> Oh, definitely. I I think all of us are vulnerable to that. And one of the the difficulties is I've even tried this with Fable where it's like, well, help me find vulnerabilities in my own code.

[00:40:59]And then it's like, well, it doesn't know, am I some rando who's looking at the code hoping to exploit it, or am I the person who's trying to maintain it and actually trying to plug the holes?

[00:41:09]>> And so I can I I appreciate the problem that Anthropic has where it's like >> you you want to give the good guys the tool and not the bad guys the tool, but how do you know which ones are the good guys and which ones are the bad guys?

[00:41:20]It's hard. And you know, Trump's answer, of course, is well, Americans are the good guys and non-Americans are the bad guys. And I don't think it's that clear.

[00:41:26]I'm pretty sure there's some bad some bad guys in America, too.

[00:41:29]>> Yeah, for sure.

[00:41:29]>> So, um that's a hard problem there. And so, without those tools, it's hard to actually prevent those tools from be able to exploit you. Now again I think you know Bitcoin and Ethereum are definitely much bigger honeypotss than XL1 is at this point.

[00:41:43]>> Hopefully in the future we're a big honeypot there as well but I think all the projects out there are going to be dealing with this problem as well. And so using AI to actually secure ourselves is going to be something which we're going to have to be very diligent about because it's it's it's a huge risk.

[00:41:59]>> Yeah. to increase the man the security of each of these blockchains. When these developers are coming together and they're having those meetings, they're going to have to have different PDFs and writeups of this is what my agent come up with this week.

[00:42:20]These are the vulnerabilities that are top priority that we need to implement changes for. These are the notsodistant future and vulnerabilities that are low low probability.

[00:42:36]>> Um so yeah, blockchains are going to need to be updating more frequently. I feel like once a month, once every two months, there's going to be updates that have to advance the same way these AI models are advancing. Every It seems like every week they're coming back and forth against each other. someone's coming out with something bigger and better and you got to try to keep up with that. But then on the flip side, when you have a software technology, you have to keep up with what AI is doing. It's so mind-blowing. Well, one thing also I think is happening is is this is the death nail in my opinion of open source because one thing we did already quite a while ago was we we stopped access to our GitHub repos for our sensitive um code which basically can be hacked or can you know can have these these >> theseabilities exposures vulnerabilities and so if if it's public and you know the worst thing you want to do is put a comment in there and say oh to-do go fix this vulnerability right it's like AI be vulnerability right there, right? And so you don't want to give that information to an AI because I'm sure these GitHub repos are getting scanned constantly.

[00:43:49]>> Oh, for sure.

[00:43:50]>> And so the first thing we did was, well, let's not give them our source code. If they want, they can go decompile what we did or whatever it is and then, >> you know, do that or from there offiscated. But there's no reason to give them an advantage, especially if it's open source and before we release it, they can even see the code that's that's there. So they can have a zero day because they saw the code >> before launched. Yeah. Right. And so you know if if nothing else you at least buy yourself some time by not giving them any access to the code until it the first day it drops it and then they have to go decompile it or they have to go and deoffiscate it and that sort of thing. So there are ways to it's kind of like you know just give yourself a little bit of a lead. So you know you always have like a twoe lead before AI catches up to you um is is nice but it's not you don't get that long of a lead because AI can decompile stuff really fast as well. So, but I think most most people out there are starting to not use open source packages because of the fact that if they know I'm using an open source package, if they find a vulnerability in that, they can exploit that in my software. So, we've been going through um and trying to strip out any unnecessary open source and you've also seen the supply chain attacks that have been happening as well, right? Axio's had that huge supply chain attack and >> which also risks for you people trying to get into your source code. So, I think it's really going to change the paradigm of how people write software.

[00:45:08]And I I don't know if Ethereum has gotten to the point yet where they've they've not had open source for their guest clients and stuff like that, but my guess is they'll have to eventually just because otherwise you're just serving it up on a platter. Yeah.

[00:45:25]Okay. Now, the next thing I want to talk about is you guys recently announced a new partnership with Theta. Can we go into that a little bit and how XL1 ties into this new Theta partnership that also includes sports teams and other big corporations that you guys are verifying the AI data? What is that about? So what we're doing is we're basically um acting act acting as a third party where we go and look at the blocks that they generate and you know go through their au inspect those and then we store um basically reports on those or or or inspections on those to our blockchain.

[00:46:11]So we make a permanent um observation on on what what we think is is appropriate for those and it depends on the type of data. Um you know we've only you scratch the surface on some of the the areas as far as well how do you validate it? So the biggest thing we do right now is we validate that it has been done right. So it's like the first question is did did somebody actually do this or not? And then the second question is well how well did they do it or did they do it the right way and so on so forth. So, you know, going deeper and deeper and deeper is part of the process, but our our first um portion of it was just validating that something actually did get done from something. And so, um you know, it's a an external third party auditing system basically.

[00:46:50]>> Okay. because I seen the Theta is working with the Houston Rockets and with the NBA playoffs just finalizing last night with the Knicks winning that was incredible. Um, the Houston Rockets are at top of mind and analytics for people who score, people who make a steal, rebounds, different things like that. Would that type of information be stored on chain?

[00:47:24]>> I'm not sure honestly what what the relationship is between the Houston Rockets and Theta. Um, so I haven't seen that data, but that would be something which they'd want to store on there potentially, I'd imagine. um or even things that are more proprietary where like >> you know times when players went for rehab or they practiced or um you know did different things so you can you can analyze those I don't know how much of it's public facing data like stats and that's also interesting where they they have like copyrights like I think I don't think you can just go and publish stats all over the place without the permission of the NBA >> because it's really it's it's a it's a reproduction of that game >> to some way >> and so like you have to get a license to actually show that. So I'm not sure if they're going to put licensable >> data in a public >> space data. So I'd imagine it might be they're either using a private storage for it, you know, or or they're putting you know other sorts of data on there which are safe to be put in public. But you know those are the kinds of things which definitely would be supported in this case.

[00:48:26]>> All right. So there's no direct correlation between the rockets X YO XL1 and Theta. Nothing in particular between the three parties.

[00:48:41]>> Well, there's particular between us and Theta and I believe there's something particular between Theta and um the Houston rockets, but our our concern is is agnostic of who their partner is. So we we like look at and observe and secure data that is theta data as far as we're concerned >> theta data that's wrong >> data yeah and where that that data came from or from which partner is for it's not something which >> it's irrelevant to XL1 you guys are just verifying the data whatever the data is >> yeah we never have code like if Houston rockets then you know >> right okay yeah no that's incredible Okay. So for verifying AI data, is that in particular to the partnership with Theta or is that something else you could do for other blockchains and networks?

[00:49:35]>> Oh, we can definitely do for other networks and blockchains and even just um AI in general without a blockchain behind it because one of the things I found interesting about AI is you ask the same question twice, you get two different answers, right? It's a it's a a probability based system. It's not a um a a constantly same system.

[00:49:53]>> And so >> storing like what what did I ask and what did I get is kind of interesting because you can't just go and reproduce it necessarily like like a hash is easy because if I hash the same payload on XL1 twice, I get the same hash.

[00:50:06]>> It's not not random at all.

[00:50:08]>> But with AI, it sometimes is. And so if I want to say well I did this and I got that it's something which is interesting which somebody has to attest to because there's no way for me to prove it's that unless the obser an observer basically recorded it at that point in time. So I think um audit logs for AI especially when we get to the point when someone says well my AI did this and the AI won't admit to it basically >> right >> you know if we have a third party observer that said well I observed the fact that you asked this and it did that you know it's it's kind of like a lawsuit right it's like the person suing and the person getting sued are kind of they're conflicted but if you have a third party who's like well I saw what happened you trust them much more Yeah, I think about like building out my I have two open claw instances and for each as I've built them out, I create different markdown files for everything I do. And even using those markdown files for individual threads, I could ask the same question three times and get three totally different answers and it pisses me off.

[00:51:19]>> I try to train it. I try to use the exact same wording between the two instances and no matter what you do, you're always going to get a different answer. The I feel like the markdown files give it direction, but it's not set in stone even with a markdown file. Have you ever experienced that?

[00:51:40]>> Oh, definitely. That that's kind of the point of AI to a large degree because it's kind of like a human, right? So if you ask the human the same question, if you ask if I ask you a question today and I ask you like how well are your your your earphones working today and I ask that same question a week ago, I might get two different answers, right?

[00:51:54]I ask the exact same question. And so you know time matters in there. Um like one thing I've done is which a lot of success is I I'll write something with claude for example and I'll say well go through and you know explain to me the architecture and you know why you made these decisions blah blah blah and why they're the best decisions. Okay, we'll do that. I take that repo that I say codeex, hey, go review this as if you're the, you know, the boss or the person who wrote it. I don't tell Codeex or Claude wrote it. It'll go through there and it'll tear it apart. It'll be like, oh, well, you know, this could be done better this way and that way and so on and so forth.

[00:52:26]>> That's right.

[00:52:26]>> And, and usually it's right. And so, actually iterating between models, I found, is actually a great way to kind of sharpen the pencil over and over and over again and get to a better and better result. um it takes a little more time and some more tokens of course, but just iterating with the same model over and over again sometimes does become feudal at a point because it kind stuck in a rut. It's a kind of like you know it's it's pot committed to the the decision that made well I made this decision I'm going with it and you know >> right that's a good point and that's something I'll definitely check out later this week is going through the different iterations between models.

[00:53:03]That was a good point because even between the two different instances when I'm kicking the same answers back and forth, even though it's two separate, they're running on the same models. So, I'm not getting an alternate answer kick back. But if I use Grock or I use uh Anthropic, there's different answers that I'll get from each of these models.

[00:53:24]And I have seen other people do that, but you mentioning that, that's definitely something I'll check out for sure. Yeah, it reminds me of the Apollo program where they um I believe they set up three computers with the Apollo program and what they always did was to have all three computers calculate the same thing and if they're all three the same obviously that's great. If two out of the three are the same then they ignore the third one. So basically it was like a little micro consensus system because they weren't sure what cosmic radiation stuff like that's going to do to the computers. So if you set up three and at least two of them um you know match then odds are that's probably going to be right. And this is kind of similar where you know if you can basically in a weird way it's a coding by consensus. If you have you know x number of of models doing something and you know most of them agree on the way to do something well that's probably the best way to do it you would think.

[00:54:10]>> Yeah that's sick. All right so to wrap up the interview today I just want to ask you one last question. Do you know or have a roundabout date of when you guys would release the XL1 quantum resistant wallet and how soon do you expect the Android and iOS versions thereafter?

[00:54:35]>> Well, the code for the wallet is already available in our SDK. Um, so if you wanted to like generate a wallet u in in memory in code and actually be able to sign things, it's there. Um, you can't send a transaction to our chain yet with that. Um, >> okay.

[00:54:51]>> Well, you probably can, but it won't work. It'll reject it because it doesn't understand it. Um, >> okay.

[00:54:56]>> So, it won't get past the meool. Um, so in the next probably four to six weeks, we'll have um all those available, the iOS, Android, and um Chrome plug-in versions of the wallets with that there.

[00:55:09]and then um shortly thereafter we'll have the ability for our chain to actually accept those.

[00:55:16]>> Okay, great. Now, what are you most excited about when it comes to building something that is quantum resistant?

[00:55:29]>> One of the things that I I've I feel about blockchain in general is the idea of permanence. And so like people like to own things that are permanent. For example, if I own an NFT, I feel it's permanent, right? It's like I own this forever. I don't have to I don't have to pay tax on it. Like if you own a house in the US at least, you pay property tax on an ongoing basis. So you're kind of leasing it from the government. If you stop paying your property taxes, they take it away. Right.

[00:55:52]>> Right.

[00:55:53]>> So home ownership in the US is not like absolute ownership, but like like ownership of NFT is like absolute ownership. like I can put this thing in my wallet, print out the the phrase for it, stick it in my safe, or for that reason burn it, and no one can ever get it. But it's it's it feels very permanent. So people want everything which they do on chain to feel very permanent. And as soon as that gets taken away um it kind of breaks the emotional benefit of blockchain like you don't feel as permanent anymore. Now I got to you know go and convince somebody trying to convince somebody to transfer money from a wallet that I don't have the key for anymore. Well, I got convince the powers to be or whatever it is. But it's like the sovereign nature of I have the private key, I can move whatever I want is very important. And I think right now people still feel that way about our legacy signature is that if I if I have a NFT that's residing in a legacy wallet with a legacy signature, I have that permanently. But I think people are starting to realize that if you were to take that and you know bury it or put it in the safe for 20 years, if it's valuable enough, someone could take a quantum computer and take it from you. And that's the one thing which people don't want. They don't want to feel that in in their lifetime at least it's possible for someone to forcefully take something from them which is onchain. And that's what I think is going to happen. And so being able to restore that comfort to people with quantum resistant um wallets and quantum resistant encryption I think is something which is very important not only for XL1 and for XYO but for the the whole mythos of of blockchain and and crypto.

[00:57:30]>> You just blew my mind because there was so much there when we were talking about the home aspect in itself. Um, a couple years ago I paid off my house and at first I was excited. Yes, I own this.

[00:57:46]But then tax bills came and I was like, I don't own this. This is crazy. I want my money back. But then as you went on in describing that, I I started to get those feelings, the negative about it.

[00:58:01]But then I thought about, okay, at least I own it. Now on the flip side, if you even if we project it five or 10 years down the road, if someone had the ability to have a quantum computer, would they be able to alter the documents and take my house the same way they could crack a blockchain to where someone could use a quantum computer and take my house? That would be mindblowing.

[00:58:30]>> It's not impossible, right? It's like depends on how you know your state like stores its its information and its data.

[00:58:38]So >> that is not good.

[00:58:39]>> It's you have to worry about all those things. It's kind of like the deep fake problem, right? You have a problem right now where >> I see stuff on online all the time where like my wife will be like, "Hey, did so and so just say this?" It's like, "I have no idea." Like that could have been a video of the person saying it. It could have been a complete lie and a deep fake. They It's really hard to tell. You know, sometimes you'll be able to see artifact here or there. and they look really real. I mean, they look like they're official documents and that sort of thing. I'm always kind of surprised that the postal service doesn't crack down on that, but um another issue completely, but um right, >> you know, people are already doing things that simply. So, getting to the point of forging documents, I'm not even sure it's actually quantum computers. I think generative AI is actually a bigger concern for that than than than quantum computers.

[00:59:24]>> Holy smokes. So yeah, all of that getting to this point of being I think I'm 38 or 39. I don't even know how old I am no more. But by owning a house at this age, I thought that was incredible. Like I I surpassed what my parents were able to do. I was happy. I was I I felt celebratory about owning the house. But then when you think about those type of vulnerabilities, it's like, man, would I want to have something on chain that has the potential to be quantum resistant in the future versus having a physical asset that could still be taken away if their computers, if their software is not secure, there's an ability for somebody to take it from you. And that blows my mind.

[01:00:19]>> Yeah, that's that's that's the hard thing. One of the things we we built a little while ago is called digital artifact which is basically you can take any any artifact whether it's a image or a JPEG or a PDF for that reason and you can put it on chain and you can you know you can basically prove that you had it at this point in time.

[01:00:35]>> Yeah.

[01:00:36]>> Now so you can take your title for example scan your title put it on chain.

[01:00:39]Now the problem is if somebody else does the same thing with a fake title and puts it on chain like well mine my onchain is newer than yours. And so it doesn't solve all the problems but it actually solves the the opposite problem where like for a copyright for example if you write something that you want to own you know a copyright on if you're the first person to put it on chain it's kind of like a way back machine you can say well that date I had it and if someone in the future says well I you know they put it on chain it doesn't matter because mine's I can prove mine was >> so you can't pro you can't prove the latest but you can prove the earliest pretty easily with our digital artifact system which solves some of the problems at least Those are the kinds of issues that you have to figure out how do you actually you know use these technologies to be able to protect yourself against those and it's quite difficult.

[01:01:24]>> This has been a great conversation. I appreciate your time today. This has been XYO layer 1 quantum AI wallets with Arri Tro. Thank you for being here brother.

[01:01:36]>> Thank you very much for having me.

[01:01:38]Appreciate it.

[01:01:39]>> Have a good day.

[01:01:39]>> You too.

[01:01:40]>> Byebye. I've been trading crypto, crypto, crypto. I've been trying to get more every time he dips low. I've been trading crypto, crypto, crypto. I've been trying to get more every time he dips low.