Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
#36 - Verifiable decentralised compute and RWA tokenisation with Blane Sims
194 Plays28 days ago
Blane is the Chief Product Officer at Truebit. He focuses on building verifiable compute solutions that bring greater trust to blockchain-based systems.
The industry has produced solutions for scaling, tokenizing assets, and integrating artificial intelligence, but ensuring security and authenticity remains a core challenge. Blane from Truebit discusses how verification layers and decentralized compute can bolster trust, reduce risk, and streamline complex automated transactions.
Securing Data Beyond the Blockchain
The conversation begins with a reminder that high-value crypto participants can fall victim to scams and phishing attempts. These incidents, often driven by social engineering and fraudulent requests for signatures, highlight a lingering issue: while the blockchain ledger is secure, the pathways that feed it information and instructions can be compromised. Many problems arise at the off-chain interface—where external data and code feed into smart contracts.
Traditional bridging and cross-chain operations, as well as interactions with external APIs and off-chain services, have repeatedly proven vulnerable. To address this, Blane explains that verification technology can certify both the data sources and the processes involved. By revealing how inputs are obtained and transformed before they reach the blockchain, solutions like Truebit’s verification layer help ensure that the final on-chain state reflects an untampered series of off-chain computations.
By using a decentralized network of nodes and game-theoretic mechanisms, verification platforms can spot discrepancies in computation results. As a result, developers gain a higher level of assurance without giving up the flexibility or complexity of off-chain workflows. This unlocks a future where Web2 and Web3 integration is safer and less reliant on trusting a single service provider.
Tokenization and Real-World Assets
Tokenization of real-world assets—stablecoins, commodities, even automobiles—has become a major trend, promising more efficient, accessible markets. Yet tokenization only works if participants trust that the underlying assets are truly what they claim to be. Without a verification layer, someone might claim reserves that do not exist or introduce false data that skews pricing or eligibility for trades.
Verification tools help confirm off-chain data, such as proof-of-reserves or asset conditions, before they enter the blockchain. This can extend from stablecoins backed by fiat to tokenized titles for vehicles or tokenized commodities. As processes shift toward automated AI-driven decision-making, having a reliable record of every off-chain step is critical.
The promise of automated workflows—like instructing an AI agent to purchase a car and finalize title transfers in a single afternoon—relies on complex webs of off-chain and on-chain interactions.
Several projects in closed beta have demonstrated the applicability of verification layers. Supply chain initiatives have used Truebit’s approach to ensure data integrity across multiple ledgers, and environmental projects leverage it to confirm the authenticity of carbon credit information. Similarly, financial tools build confidence by verifying the logic behind credit checks, stablecoin reserves, or complex remittances. All these examples show that, beyond theory, verifiable off-chain computation can underpin a multitude of real-world use cases.
This podcast is fuelled by algorithmic cryptocurrency trading bot platform Aesir.
Recommended
Transcript
Intro
Crypto Hacks and Exploits
00:00:09
cyberpunkmetalhead
One thing that I meant to mention is if you're following at all crypto hacks, it's one of those like things that I kind of enjoy reading about because I enjoy kind of the like first of all, how creative some people can get. And then second of all, you know just the sheer amount of like exploits that are still you know discovered on almost a weekly basis. like I found this article about hacks or prominent hacks in 2024 and there's literally probably like almost one or two every month, ah like clockwork. And the most recent one is there's this um address that was attributed to an ETH whale that got drained with over $35 million um dollars in ETH due to being fished into signing a fraudulent like a a scam, a signed transaction that he shouldn't have, he shouldn't have signed.
00:01:04
cyberpunkmetalhead
And that just got me thinking, you know, we're in 2024. It's been so many years since we had smart contracts and we still have, you know, so many vulnerabilities. And I just just thought it's it's kind of wild that we're still at that stage.
00:01:19
Blane Sims
Yeah, for sure. I mean, that's you know that that type of tax sounds like it's almost social engineering more than anything, the the type of fishing that you know that that takes place every day. so But the layers of vulnerabilities are are yeah certainly kind of you know outside of the Yeah, outside of the the realm of even even what every, you know, any person can
Phishing and Smart Contract Vulnerabilities
00:01:48
Blane Sims
consciously think of. So I agree.
00:01:51
cyberpunkmetalhead
Yeah. um Well, just to get a line from the article, it says, just just reading here, experts have explained that the phishing scheme involved deceiving the whale into signing an offline permit message, a common practice for DeFi for authorizing tokens transfers without revealing private keys. Once signed, the attacker exploited this permit to steal um the the wrapped ETH tokens.
00:02:13
cyberpunkmetalhead
um which is ah yeah it It sounds like, again, a protocol similar to a bridge protocol, a bridging protocol, um which I guess historically has been one of the infamous sources of of exploits in the past.
00:02:22
Blane Sims
Right.
Bridge Protocol Concerns
00:02:28
cyberpunkmetalhead
um and We're still yet to see like emerging i guess emerging technologies that are that really you know stopping that. like As long as there's a bridge involved, I feel like there's always going to be a certain amount of of hacks or exploits that are going to be discovered sooner or later.
00:02:45
Blane Sims
Well, I think the you know I think that's ah that's a really interesting point where um you know the concept of of bridging or or you know trusting an intermediate technology blindly to um you know to do what it's what it says it's supposed to do um is is rife with you know the opportunity for things to go wrong. And I think that's where you know true bit and the the idea of verification, the idea of having complete transparency into you know how, whether it's a a bridging transaction or
00:03:23
Blane Sims
you know in a much broader sense, off-chain you know data, so much of of the the transactions that we're doing on-chain, rely on off-chain data, um having the the entirety of those transactions transparent and verified, ah you know we see as really the the only remedy to on you know being taken advantage of and in using verification as a yeah as a layer of inherent trust um because you you can always trace back and you know see what is supposed to happen before it happens and also see how you know data transfers, um you know bridging transactions, et cetera, are executed in the moment. um you know that's That's critical to it's a moving beyond this you know this early stage of
00:04:19
Blane Sims
um you know really easy attacks on everyday transactions in the ecosystem.
Introduction to Truebit and Ethereum Scaling
00:04:27
cyberpunkmetalhead
Right. So your your work with Truby, it revolves a a lot around that, doesn't it? Like kind of finding a safe and transparent way of dealing with with that. So maybe you want to, I think this is a good point, a good time to kind of get ah a little bit into the product that you're working on and and like how it came about, but also your overall and your wider experience in crypto and blockchain and like how you got here. I think that would be interesting.
00:04:52
Blane Sims
Yeah, absolutely. um So yeah, so Trubit, Trubit's a really interesting project, really interesting company. Trubit's been around ah really since the early days of Ethereum. Trubit started as a research project. Our founder is credited with writing one of the first academic white papers talking about how to securely scale Ethereum transactions and really has laid the kind of the foundational kind of research around verification of off-chain compute.
00:05:22
Blane Sims
Fast forward today, Truebit is creating what we see as a verification layer ah that secures off-chain data, which is the data that we see is being really behind every on-chain transaction. So app developers use Truebit to certify data that they source from real-world systems, maybe through APIs, to safely move data um between different chains, and to transparently automate complex transactions.
00:05:49
Blane Sims
Truebit is a form of decentralized compute, so we have a decentralized compute network that provides immutable proof of authenticity for the critical information, algorithms, network connections, all these things that are leveraged by financial transactions, other high-stakes transactions on-chain. I've been involved with Truebit for a little over two and a half years now.
00:06:12
Blane Sims
um I come not from the world of blockchain, I come from the world of enterprise compute. So you know as and as ah kind of an infrastructure are you know ah person,
00:06:25
Blane Sims
I really see in you know as a joint group at the opportunity to help shape ah you know kind of the next 20 or 30 years of of compute from a very, very different standpoint where yes, it's you know anchored around blockchain transactions, but All of the surrounding um you know work that goes on, the automation work, the AI work, the the integration work that's driving these transactions needs to be safe and secure ah to a degree, which that's really not possible ah today and until the emergence of Trubit and other verification technologies.
Off-chain Compute and Data Integrity
00:07:02
cyberpunkmetalhead
So you mentioned something interesting when when you said off-chain compute and I wanted to get a little bit more into that um and exactly what it is that you mean by off-chain compute and ah what proportion of compute happens or should happen off-chain versus on-chain.
00:07:05
Blane Sims
ah
00:07:20
Blane Sims
Yeah, absolutely. I think this is a really great point. on Off-chain compute, I think from some really early definitions, really just referred to anything that was happening off a particular chain. So yeah it was you know relative to Ethereum. Anything that wasn't happening on Ethereum um was considered off-chain.
00:07:39
Blane Sims
uh, even if it was happening on another, another blockchain, right? I think we take a much broader definition of option commute, uh, where we see, uh, the, uh, kind of the entirety of processes that certainly today don't fit within the constraints of, uh, of what's possible or at least, uh, practical on chain. Uh, but, you know, I think, you know, that we're, you know,
00:08:05
Blane Sims
10 plus years into ah the you know kind of the the decentralized era, um we we don't think that it's a safe assumption to say that over time, all compute will be on-chain compute. There will always be legacy systems. There will always be um types of compute that are written for reasons of privacy or reasons of practicality in terms of the complexity of what's going on.
00:08:31
Blane Sims
are better handled not within the construct of a of a ledger. um So we we look at off-chain computers being the really the 99% of code ah that makes up a you know decentralized app. That's not code that's running in the smart contract or code that's that's ah within um you know within the blockchain itself.
00:08:52
Blane Sims
ah Yes, that does include some of the bridging code or the the code that needs that needs to be there to move data between blockchains. ah But it really you even more broadly centers around um the you the the ability to integrate with with all the real world systems that are out there. Something like 200 million Web2 APIs in existence today, those are not all going to go.
00:09:18
Blane Sims
on non-shade in any short timeframe. So as you would want to reach out or need to reach out into those systems to pull data in, how do you make that data ah reliable and provably reliable?
00:09:32
Blane Sims
I think that's the base layer of the verification that we're doing. But I think when you kind of keep peeling the onion back, which this is really a multi-layer problem, you you try to get back to the core of where
00:09:43
cyberpunkmetalhead
Hmm.
00:09:46
Blane Sims
Where does the data come from? Is the data reliable? um You see ah you know AI and machine learning models that are increasingly important to to transactions that are taking place on chain, many of them having impact but perhaps in real time. um so That's a you know essentially an attack vector, that that verification.
00:10:08
Blane Sims
and ah you know can can play a role and in projecting.
Tokenization and Real-world Asset Verification
00:10:13
Blane Sims
You see ah the um ah you know the ability to ah connect and move um you know information and in automated ways becoming really important. And then as we we're starting to look at you know new applications as the world of of you know asset tokenization starts to take off. um You see all of these latent, ah you know very, very trust-based, intermediary-based processes that that drive kind of the financial world today that are yeah you know kind of trapping or holding back the ability of those assets to tokenize effectively. um And the the question is, how do you automate ah you know things like proof of reserves? How do you automate, you know if I say I have
00:11:01
Blane Sims
$100 million dollars of gold in reserve. How do you make sure that that's actually true? um How do you you know move beyond you know even some of the kind of the the basic financial account proving into kind of ah really even checking in real time?
00:11:15
Blane Sims
um, how, uh, you know, underlying assets, you know, in the the world of commodities, for example, uh, you know, what the condition of, uh, of crops might be that are, uh, that are tokenized and available for sale. So it's really like a long view of, you know, we start today with, uh, you know, these integration points, but you, you got to keep killing that, with that onion back, uh, farther and farther do you get down to, you know, where does, uh, where does our, uh, our, uh, kind of ground truth originate and how, how does that made secure and, uh, improve a true
00:11:46
cyberpunkmetalhead
Yeah, no, that that is a very good point. So I think it's my follow-up question to that would be, let's say that you have um a an application that is mostly a Web2 application, but just does fetch some Web3
Integrating Web2 and Web3: Challenges and Solutions
00:11:59
cyberpunkmetalhead
data.
00:12:00
cyberpunkmetalhead
But for all intents and purposes, it's it's traditional website Web2. I just go and fetch some NFT information to display on my website and then let people like Mindy NFT if they want to.
00:12:04
Blane Sims
yeah
00:12:09
cyberpunkmetalhead
Would that need to then have a Web3 counterpart? Would it have to exist on Web3? Would it have to have a proof that something has happened apart from obviously the the on-chain actions that that are record recorded on the ledger?
00:12:24
Blane Sims
Yeah, yeah, that's an interesting point. so So from the standpoint of Web 2 reaching into Web 3, it's a little bit of a different story in some ways, right? um I think maybe we can go back to the example of you know that that unfortunate fraud that took place that you were talking about in a minute to kind of...
00:12:34
cyberpunkmetalhead
Hmm.
00:12:45
Blane Sims
illustrate this, but in in in a world where I have built essentially a Web2 application um that's going to call into Web3, the critical point for verification there isn't actually necessarily in retrieving the Web3 data. I think that's um you know that's something that is already kind of fundamental fundamentally verifiable because you can go and see if I'm retrieving data back that it's not the data that existed on chain. It's really in reverse when I'm going to take data from my web2 application and now I'm going to use it to mint that NFT or to to create that token or to to transact. um
00:13:20
Blane Sims
was anything altered ah in that in that path from um you know from the origination of that data um before it actually you know gets committed in that transaction is really the way that you that we see that kind of taking place. I think the reverse that you're talking, ah you know the to to the fraud question that you brought up earlier though, there is an interesting point to to kind of think about in reverse, which is um If you're being presented within a Web2 interface,
00:13:51
Blane Sims
um you know a construct that you're but you're about to transact based on a you know construct. How do you prove that the construct, and I think you were saying that this was a bridging technology that was the source of that hack and someone had misrepresented you know the bridging technology is is was what I was hearing. How do you make sure that you're actually seeing what you're being told? And that that that is a place where verification can actually come into play and and create that you know that safety layer.
00:14:21
cyberpunkmetalhead
Yeah, so you know how in Web3 and blockchain, there's the word abstraction layer is just being thrown around a lot recently and the idea of you know making making it easy for people to onboard and easy for people to understand to the point where they they don't have to have a certain specialized knowledge to use Web3.
00:14:28
Blane Sims
Yeah.
00:14:39
cyberpunkmetalhead
And you know I'm a big believer that that's where we're going. But I also recognize the risks of onboarding people who have no idea of what the risks could be potentially because the the easier it is to use, then the easier it is to potentially fall prey to some to some bad actor when you don't understand what's what's going on.
00:14:52
Blane Sims
Sure.
00:15:02
cyberpunkmetalhead
So I'm just wondering. Is there a sweet spot there? Do we want you know Web 3 to be the easiest thing to use in the world? Or is it going to naturally have a bit of a learning barrier to to prevent people from from making mistakes, that you know working with a technology that they they don't really understand under the hood that will?
Balancing Web3 Usability and Security
00:15:24
Blane Sims
Well, you know, I think when you're, when we're in a world where what one would assume is it even a relatively sophisticated web through user, the, you know, the, the, the, the eth whale that had 30 some odd million of tokens drained from their account, uh, because they misunderstood the context of the transaction that they signed.
00:15:37
cyberpunkmetalhead
Right. yeah
00:15:45
Blane Sims
Um, I think we, we really. you know if the If the experts or near experts can't get it right, I i think we have to solve the usability problem for ah you know for the everyday person um who ah you know quite frankly you know probably doesn't fundamentally understand what a wallet is, how cryptography works, how important it is, and how to even keep You know the details of their wallet safe. I mean, I think they you get over a bit of a learning curve as you step in uh, you know step into web3 Uh, you know become token holder, etc. But um, so this this point of abstraction back to your question then um, yes, huge you know huge step forward, uh for the uh, you know the scope and the applicability of of the yeah, you know the yeah applications we're building in web3 um, but also
00:16:44
Blane Sims
opening new vulnerabilities. And, um, and so what, what needs to, you know, fill in that gap are additional, uh, kind of checks and balances, um, verification, uh, provides us checks and balances. Um, you know, there needs, you know, being able to expose back to, I guess I'm just going to ask you, how do you expose back to, uh, you know, an everyday user, um, that there.
00:17:08
Blane Sims
uh you know their transactions are verified or this environment that they're doing transactions is a verified environment um you know it's uh if users don't understand um uh you know kind of how to you know the basic fundamentals of wallet they're unlikely to understand how to uh go and run verification on a cryptological proof right uh which is the ultimate kind of output of of a system like truebit um but that is a very
00:17:30
cyberpunkmetalhead
Right.
00:17:37
Blane Sims
um small problem compared to, okay, is this a safe environment ah to work in?
00:17:43
cyberpunkmetalhead
Right. yeah and and that And that's a fair point, yes. I feel like definitely the the safety aspect of it is is arguably the most important one here. um I feel it's, in a sense, it's natural with every emerging technology. that Innovation happens to some extent you know at the expense of of the early adopters trying to figure out what how to actually use it.
00:18:08
cyberpunkmetalhead
The more you get exposed to it, the more you use it, it's inevitable that a small percentage of the user base will be targeted by by hacks, by exploits.
00:18:08
Blane Sims
Right.
00:18:16
cyberpunkmetalhead
It's just, I feel like it's just the nature of this piece, which is innovation, right?
00:18:17
Blane Sims
Sure.
00:18:21
cyberpunkmetalhead
It's in a way, you know, like Salut to all the soldiers that have had experiences in the past, which allowed us to to grow this industry to where it is today. So one other interesting um aspect of Trubit that I wanted to get into a bit more is you've mentioned automat ah to automate complex transactions.
AI-driven Automation in Tokenized Transactions
00:18:41
cyberpunkmetalhead
um What do you mean by that exactly?
00:18:45
Blane Sims
Yeah, great great question. So um so think about um you know the the world that we're entering today, which is increasingly a world where um you know our real world assets are are being know represented on chain through tokenization.
00:19:05
Blane Sims
you know, huge, uh, movement, excuse me, within the last year, uh, of, uh, you know, billions of dollars of, uh, of assets, uh, you know, now present on chain. And then, uh, uh, you know, uh, potential, you know, estimates range anywhere from call it a two trillion to up to $30 trillion of assets over the next 10 to 15 years, uh, that, um, you know, that can be,
00:19:31
Blane Sims
ah you know, represented on change through tokenization. So this this tokenization world is happening on one hand. And on the other hand, we have ah the world of AI, right?
00:19:43
Blane Sims
Where increasingly um we're and in a space where humans aren't necessarily going to be the best answer four you know for orchestrating and in kind of making our workflows, everyday workflows.
00:19:44
cyberpunkmetalhead
Mm.
00:20:01
Blane Sims
uh, kind of, uh, happen. It's, it's going to increasingly be driven driven by AI. So this is a framework going against which, okay, we're wanting to move more and more value on chain. ah to unlock liquidity, to um you know increase access to different types of of investments, to much broader investor base, ah to make the financial kind of systems that are you know that dealing with assets much more efficient. And then this other kind of huge efficiency ah storm from from AI, with that what that breaks down to from our perspective is that's a lot of automation that needs to take place. So if I you know take a
00:20:36
Blane Sims
um you know kind of We recently went through a kind of discussion with a group where we we walked through what would it take to, for example, completely automate purchasing a car.
00:20:47
cyberpunkmetalhead
Heh.
00:20:48
Blane Sims
Were you asking AI chatbot, hey, um You know, this is the type of car I'm looking for. um You know, maybe you start with, you don't know what you're looking for. You're looking for an electric SUV or something that of that sort. um This is the amount of money that I have to spend, you know, and I want it to be even on a new car, but a used car, et cetera. All these details you give to the chat bot and you say, go figure this out. And I want the car here by 5 PM today. I mean, it's a little bit of a silly thought today because, you know, you wouldn't really be able to do all of that in today's world, but with the,
00:21:19
Blane Sims
kind of all of the Web2 systems that are coming online that are already there in many cases to to of allow for this automation from an API standpoint with the tokenization of auto loans.
00:21:33
Blane Sims
Um, with the tokenization of auto titles, right? Which is actually starting to happen. You can see in kind of the state of California starting to, uh, you know, actually adopt, um, blockchain based auto titles.
00:21:45
Blane Sims
Um, you know, this becomes a possibility.
00:21:47
cyberpunkmetalhead
Oh, yeah.
00:21:48
Blane Sims
Um, how on earth would you have all of those different transactions, all of those different components working reliably, um, if there weren't safeguards to make sure that every every component along the way is is is doing what it says it's going to do. And then more importantly, to to make these things go at the speed that we want to go, um a lot of the you know in that world today of buying an auto, a lot of the ah the transaction is actually gated by human verification. Some person is actually checking to see, do you have money in your bank account? Do you actually have a credit score that that warrants this type of loan? is this you know Was this car actually
00:22:27
Blane Sims
um you know, totaled or not before you got it.
00:22:30
cyberpunkmetalhead
Right.
00:22:30
Blane Sims
um As as the the space for that human verification starts to shrink because the timelines are shrinking, you know, that's where automation really drives the need for, you know, for the type of verification that treatment does.
00:22:44
cyberpunkmetalhead
Right, yeah. Yeah, that's that's a really good example. I could, I mean, any any, I feel, you know, all right, if developer could probably spin up ah like a really crude prototype of that app in a weekend, um if they had a bit of time in their hands, you just need, okay, you need some source of liquidity, you could go to some liquidity pools, get that liquidity and alone, turn it into fiat, convert it to fiat, connect to some kind of,
00:23:05
Blane Sims
Yeah.
00:23:11
cyberpunkmetalhead
um car dealership API, get an AI in between to understand your commands as prompts, and then execute that with with calling the functions, get the AI to call the functions, and you could have a car you know right at the front porch before the end of the day.
00:23:27
Blane Sims
that's That's fair, which is amazing. But at the same time, are you going to trust the car that shows up? And are you goingnna you know are ha and and you you mentioned the car dealer in this transaction. that The car dealer is that is one of those one of those points of human trust. Does the car dealer is Is making sure you know that they and and even imperfectly making sure because card users do still get scammed you know does the person that's there find a car from. um If it's a used car transaction does does that person even have title to the car to sell it to them and are they going to hold and wait for the title to clear no because they might want to sell the car right away.
00:24:04
Blane Sims
um And, you know, title clearance can take months, to be honest with you, as you're sending things off in the mail to, you know, different state authorities, et cetera, to do that. So, um so yeah, exactly.
00:24:14
Blane Sims
You kind of make the point. The technology is there to do that today. You know, you could, you could, you could emulate or simulate this through chat GPT.
00:24:18
cyberpunkmetalhead
Yeah.
00:24:21
Blane Sims
I've actually done it. You can even get chat GPT to suggest some of the API code you needed to write to make it happen. But do you fundamentally you know, want that car that's going to show up at your door at the end of the day and feel like it's not going to get taken back, you know, no no take back seats, right?
00:24:38
Blane Sims
um Because of, you know, some error in the transactions.
00:24:39
cyberpunkmetalhead
yeah
00:24:43
Blane Sims
That's, you know, I think that's one example of where things are going. So cars are, you know, cars are one thing, but, you know, even before we get to that fully automated world, ah you know as we Again, we go back to where our ah where are things starting from a real world asset tokenization perspective. We're not starting with cars per per se. Maybe some people are experimenting there, but the you know that trillions of dollars is coming from the kind of the top of the capital structure. right um and um so you know The starting port point you know with stablecoins being kind of the largest asset class there, it hits this verification problem immediately. and It's only partially solved right now.
00:25:22
Blane Sims
like Do you have the reserves that you say you have available that are backing the value of the
Verification in Off-chain and On-chain Transactions
00:25:29
Blane Sims
stable point? Or if you're moving into and kind of treasury bonds, which is the next category, where are those bonds at? um And can I ah do something other than just trust you ah to say those things are placed when I can't actually see those things because they're off chain? How do I know and verify that they're there? That's the um that's the problem statement against which you know, off-chain verification, off-chain compute has to operate, right?
00:25:56
Blane Sims
We need to see that gap between, okay, you say you've got something which honestly, you know, even the most trustworthy players prove themselves untrustworthy, you know, sometimes by accident, sometimes not, you know, if we want to really
00:25:57
cyberpunkmetalhead
Yeah.
00:26:11
Blane Sims
get the you know the the true liquidity, the true movement of of these assets going, the the friction that's there because of lack of information really can only be solved through verification.
00:26:25
cyberpunkmetalhead
Oh yeah, 100%. And I feel like that was kind of one of the experiments, albeit a failed experiment, but with the algorithmic stablecoins like Luna, right? and And the Luna's own stablecoin.
00:26:35
Blane Sims
Yeah.
00:26:36
cyberpunkmetalhead
It's okay, we're not going to back it with us with the US dollar. We're going to create this like bucket mechanism that's going to rebalance, you know, a ah burn mint mechanism between Luna and the stablecoin.
00:26:52
cyberpunkmetalhead
That's going to dynamically rebalancing incentivize people to just you know burn or mint depending on on the supply and demand which on paper it's a fantastic idea i just i love the concept of having you know a stable coin that's not packed by a fiat currency.
00:27:10
cyberpunkmetalhead
I think that's like the ideal standard from from the perspective of Web3, but I feel like it's it's almost impossible to pull off because these things that spiral all the time and there's just no no sense of of actually of of tangible value.
00:27:26
cyberpunkmetalhead
um And I feel like it's ah it's a difficult one to to to solve. But then at the same time, you're looking at the you're looking at the US dollar, there's ah there's an infinite supply. There's no cap on that.
00:27:37
cyberpunkmetalhead
like What actually keeps its value is is its value in the market you know against all the other global markets.
00:27:37
Blane Sims
yeah
00:27:44
cyberpunkmetalhead
So it is stable it's a stablecoin is stable if it's backed by US dollar, but it's not stable in the actual sense outside of the US dollar environment in ah in a sense.
Blockchain's Role in Global Finance
00:27:57
Blane Sims
Yeah, fair fair enough. so i you i think that's the Again, I think that's the the kind of cusp of change that you know that we're we're at right now, that era of experimentation of of really seeing the blockchain, the crypto universe being a world unto itself. A lot of, to your point, great experiments that didn't work out, unfortunately, to the tune of, some people lost a lot of money when those experiments didn't work out. If we're asking the world to take on this incredible technology infrastructure that we've created,
00:28:36
Blane Sims
ah to become the you know the underlying rails for the you know the global economy to operate on, which is what we're starting to do and starting to see a desire. Maybe it's early, but starting to see a desire of of kind of the and of existing financial mechanisms to adopt this this technology because of all the kind of really difficult problems it solves for existing you know kind of and financial applications, um then we can't you know We can't have it be a world where, you know, proof things just go away overnight, right? I think everyone knows that. I think there's, you know, majority of the work that that we see going on in the and the infrastructure space around, you know, regardless of what blockchain you're working on is around, you know, security, privacy, um you know, data availability, all these core foundational things. And verification is one of the, you know, we see one of the pillars of of all of that work.
00:29:33
cyberpunkmetalhead
Oh, yeah, it's it's just so important. 100%. What are your thoughts on Ripple? By the way, you've mentioned underlying layer for global currency. And if you talk to any Ripple maximalist, they're going to tell you, oh, it's going to be Ripple. Like, Ripple's going to be the coin that all the banks use and and and that everyone, kind of all the financial systems are are eventually going to embed themselves into and and make use of Ripple to transact globally.
00:29:59
Blane Sims
I think that sounds great for Ripple um and for the folks that hold hold their token. and um I don't personally make prognostications around any and particular and token or technology because I think there's no way of knowing today ah because again, we're we've we've kind of created a little bit of our own universe here and we've created some really interesting experiments to to your earlier point for how things might work, but does any one technology become the it ah You know the e foundation my I guess perspective is no that there is such a ah variety of of technology that the the strength of of watching the strength of crypto is actually
00:30:41
Blane Sims
how resilient it is because there are so so many different ah you know chains, there are so many different you know node operators that are securing the chain, staking into those chains, that it's kind of the entirety of that that makes, you know ultimately makes things safe versus that's all piled into, you know nothing gets rippled into a great success story in its current form. but um But yeah, I think whether it's, you know you know ripple on one hand or any technology, any platform. So it's so hard to say that that's the you know that's the the one that's going to play things out.
00:31:17
cyberpunkmetalhead
Oh, agreed. ah Definitely. I feel like that that does apply to any technology and and any especially any any blockchain, because each blockchain has its own you know level of like ultimate believers, like maximalists who believe like this is this is it.
00:31:33
cyberpunkmetalhead
And if you're just a bit, you know, if you just detach yourself a little bit from that and you you you look at all these different communities, you know, like Bitcoin Max is saying Bitcoin is the best, ETH Max is saying it's the best, Solana is like, well, probably like, you don't know, like maybe none of them are.
00:31:33
Blane Sims
Yeah.
00:31:44
Blane Sims
yeah
00:31:47
cyberpunkmetalhead
maybe Maybe they've all feel, you know, maybe they're all going to communicate with each other in this abstracted layer and it's not going to matter which which coin is which, or maybe there's going to be a completely new solution that's going to come out.
00:31:58
cyberpunkmetalhead
um I just, yeah, I just think it's and it's It's great to to have that conviction, especially if you're finite. I feel like if you're also financially invested, it's hard to detach yourself. like Most people, they they tend to kind of you know really believe in that thing because they they put their money where their mouth is. But you also have to accept the fact that it might not you know there's there's space for more than just one. I don't think there's just one unique coin. I just thought since you mentioned like global financial system, I thought if
00:32:29
Blane Sims
Yeah.
00:32:30
cyberpunkmetalhead
if if if ripple is something that often comes to mind ah in in your case.
00:32:36
Blane Sims
um Yeah, no, all fair, all fair. I mean, Ripple certainly comes to mind, but again, among among others, so.
00:32:44
cyberpunkmetalhead
Yeah, fair fair point. um So let me go back a bit to compute and some of the things that Truebit provides in in terms of in terms of like
Truebit's Decentralized Compute vs AWS Lambda
00:32:54
cyberpunkmetalhead
functionality.
00:32:54
cyberpunkmetalhead
So I thought it's really interesting and I wanted to get a bit more here, a bit more from you. So you've said that obviously it's a decentralized computing platform. So in a sense, like AWS but on decentralized networks, right?
00:33:08
cyberpunkmetalhead
On on a decentralized network.
00:33:08
Blane Sims
Yeah, I think that's the you know that's the always the but starting point comparison. so um So to be more specific, you know within AWS, AWS is a huge ah huge world, right? um But if you want to kind of specifically compare kind of a starting point for Truebit,
00:33:28
Blane Sims
Trubit operates a bit like AWS Lambda. So AWS Lambda is what's called serverless compute.
00:33:33
cyberpunkmetalhead
Mm hmm.
00:33:36
Blane Sims
So you, you as the author of of you know some some code, don't have to worry about any of the of the particulars around how that code executes.
00:33:47
Blane Sims
You just ah submit your code on the AWS side of AWS Lambda and it magically works. Same thing with Trubit, Trubit verifies our compute platform. on So developers will write code in languages like JavaScript or Python or Rust or C++ plus plus ah or will write API calls that can be made through through a platform. And our platform manages ah the the execution of those those calls. And the decentralized component is then where
00:34:20
Blane Sims
you know we see is that layer of security. right it's not You're not trusting Trubit, like you have to trust AWS ah to go run these calls. um And more importantly, the um you know the the data that these that these functions are representing um itself isn't subject to implicit trust because you're showing your work as you run uh, you know, your code through true, but you're showing your work of how you got the data of how it was transformed and, and, uh, you know, how, how the resulting data is then used inma in a transaction.
00:34:55
cyberpunkmetalhead
That's super interesting. Okay, so it's like serverless lambda functions and you can use several programming languages to do that. um can Are you able to deploy um private and public applications or is there a limit there? Is there a limitation there?
00:35:09
cyberpunkmetalhead
so
00:35:10
Blane Sims
ah Yeah, so um we support both you know a private compute network and kind of public. kind of public networks. so We see kind of the public version being the default, where um when you're looking for verification, you know, in the on-chain, off-chain world, often it's because you you very publicly want people to see, ah you know, the the the source of your data, the code you're using, and the and the outputs. But going back to, um you know, kind of this
00:35:43
Blane Sims
this new era of all sorts of different types of assets being represented on chain, um you know it might be a private equity asset you know with ah with a you know relatively small investor base that's that's being represented that still needs a verification for trust amongst those investors. That doesn't need to be public. So and and in that sense, a private network would be how we we we implement and execute that.
00:36:08
cyberpunkmetalhead
okay and Aside from the fact that it operates on decentralized nodes, there's no real distinction between Truebit and any of the other compute platform with traditional web-to-compute platforms in terms of in terms of use cases.
00:36:23
Blane Sims
Well, I think um under the hood, the distinction is huge, right? So the 2-bit verification protocol, which we call the verification game, um this again is ah you know what allows us to say and provide, and not just say, but provide proof back to yeah back to you that that this ah this compute transaction, this function call or API uh integration that you're doing uh is correct right so um you know traditional can be platforms don't do that um and uh aren't aren't you know really geared to uh to provide any any sort of proof other than
00:36:53
cyberpunkmetalhead
Right.
00:37:04
Blane Sims
you know, you, you know, the AWS says that they did what they were supposed to do. And you, as the person that wrote that code, you know, kind of have complete control over what it does. So people need to trust you, the coder, as much as they, you know, as anyone, and that's not necessarily a fair ah trust assessment.
00:37:21
cyberpunkmetalhead
Right, yeah, because it it implies implies trust is necessary.
00:37:24
Blane Sims
Yeah.
00:37:25
cyberpunkmetalhead
Yeah, but that's a that's great point. And and as ah as a node operator, um are there any requirements to become a node operator for Truebit? Can anyone apply to become a node operator?
00:37:37
cyberpunkmetalhead
like What is the relationship between Truebit and the nodes and the code running on these nodes?
00:37:42
Blane Sims
Yeah, great great great question. So we're we're in the very late stages of our of our beta right now. And so we're still working through ah some of those some of those details, I think, in in general. um you know we'd We'd be looking at the patterns around staking that we see as being a really successful pattern for ah you know for committing resources into our network.
00:38:07
Blane Sims
ah the treatment verification game itself. Again, going back to treatment search and research, this is you know in 2016, 2017 timeline, really some of the original research around how you know game theory in terms of using incentives and penalties can help ensure you know a norm of good behavior amongst all of the providers.
00:38:28
Blane Sims
So that staking yeah know function then is also drives the rewards and the penalties for for any misbehavior that might might be taking place. But all the details of you know ah who how much all of those things are still kind of being ironed out as we kind of ah wrap up our beta.
00:38:48
cyberpunkmetalhead
so well That's exciting. that's it So when are you guys planning to launch then?
00:38:50
Blane Sims
Yeah.
00:38:52
Blane Sims
um our Our plan right now is for for beta to wrap up in early Q1 of next year. So launch will come shortly thereafter.
00:39:00
cyberpunkmetalhead
and Is that close or open beta currently?
00:39:03
Blane Sims
We're in a closed beta right now. We're going to go to an open beta as the final step and then and the launch from there.
00:39:09
cyberpunkmetalhead
Yes, that thats that's very exciting. So but in terms of um obviously reward mechanism and such, and and you know however much you're allowed to speak about this at this current time, um because it's they run you've said they run serverless, right they don't run on on specific, you can't select a specific you know node or or computer to run your your code on, how do you determine which node was given the job and how much work they've put in that, how much compute they they contributed with?
00:39:33
Blane Sims
Yeah, yeah, great great question. So yeah, this is all of these details are actually at this level are available in our in our most recent technical paper that describes what we call the Truebit Unchained Protocol um that kind of talks about, you know again, how that that verification ah you know is is brought forward into the serverless environment.
00:39:47
cyberpunkmetalhead
Mhm.
00:39:56
Blane Sims
But at a high level, so nodes are randomly selected. and we um We actually look at ah kind of the blockchain you know as a source of randomness.
00:40:11
Blane Sims
So looking at kind of block IDs, things like that, that help us create a random seed to ensure that we're selecting a node randomly.
00:40:12
cyberpunkmetalhead
Right.
00:40:19
Blane Sims
There's this concept of transparency as node selection is taking place. All of the um you know all the nodes that are available for selection are are made transparent.
00:40:30
Blane Sims
the you know the algorithm that we use is transparent, et cetera, so that ah there's yeah know there's a guarantee that that we're not playing favorites. um And and in in general in the design, because there are some, you know this kind of dispatching mechanism who gets to work is an example of it. There are some kind of important functions ah that TruBit is managing as part of that. There is a ah layer of we We look at this kind of policing that's going on so that anything, so the dispatcher that's going to be giving out the work, um you know their work is audited by the nodes as well, or auditable by the nodes as well, so that you know all the all the actors in the system have the ability to kind of check and balance each other.
00:41:18
cyberpunkmetalhead
That's very cool. um
00:41:22
cyberpunkmetalhead
ah so ah Just as a follow up to that, um but you said it's random allocation of of the work. um but Could you have as a user, as an end user, could you have any or would you have any control over ah proximity or location? Let's say I want to deploy some application that is really dependent on it being proximity close to the data source to minimize the latency. ah Is that something that there's an option for me to do?
00:41:52
Blane Sims
Um, sure. So as we, you know, expand our node network globally, I think, uh, you know, assuming that there is a sufficient, sufficiently safe quorum of nodes available in a region, you know, we're going to optimize the, um, you know, the routing, uh, for, for, you know, as close to near real time response as possible. So some of that is in kind of built in to, uh, into the system. And I think, uh, you know, there's always the ability, uh, to provide metadata, uh, to kind of, uh, you know,
00:42:22
Blane Sims
ah you know express you know what you're what you're looking for as you as you dispatch a task within within the 2-bit compute network. um Some of those are the um more kind of problems for you know kind of eventual scale versus in the starting state of the of the network.
00:42:36
cyberpunkmetalhead
Right.
00:42:40
cyberpunkmetalhead
Yeah, for sure. no that that that That's fair. That's great though. um I will definitely read your white paper because I think that there's some really interesting stuff there that probably already answered some of my so my questions regarding this. um I'm just thinking in terms of like the interaction with the nodes as a user, I feel like it's it's really important and really exciting as well for me as ah as a user to understand how the nodes you know compute my data and what control I have over these nodes.
00:43:08
cyberpunkmetalhead
um So just to kind of wrap my head around this, would you then say that the compute itself is off-chain traditionally, but then it's back on-chain as the output, as the verification, but the node itself, or is it just a semantic thing?
Consensus and Data Integrity in Decentralized Networks
00:43:28
cyberpunkmetalhead
Because technically you deploy it onto a decentralized network. You could argue, it well, it's on-chain because it's on a network, isn't it?
00:43:31
Blane Sims
Yeah.
00:43:35
Blane Sims
Well, yeah.
00:43:36
cyberpunkmetalhead
And then on a ledger, I guess.
00:43:36
Blane Sims
i mean um So we we certainly don't see the decentralized network as being a blockchain. and I think there are some um some defining characteristics of of blockchains, and turn especially around ah you know the the the presence of a ledger right that du that's kind of tracking transactions and and state as
00:43:45
cyberpunkmetalhead
Right.
00:43:53
cyberpunkmetalhead
Yeah.
00:44:01
Blane Sims
you know as transactions happen over time um and the i and so forth. um That said, um what we've done again with Jason Toich, our founder, having been really present and and and and kind of you know at the vendor in some senses of of a lot of kind of the the more kind of advanced consensus mechanisms that drive ah you know scalable transactions on chain is take that same approach to um to consensus and apply it in ah in a decentralized off-chain compute network. So um so like let's kind of just maybe play forward. So you have a
00:44:45
Blane Sims
Um, you have a, uh, you know, again, this idea of ah of a Lambda function, uh, that's, uh, executed by true bit. So you make a call to your function. Um, that call is dispatched, uh, randomly to, uh,
00:45:01
Blane Sims
right now at least two nodes. ah So the verification that we're doing is we're kind of competing, you know, pitting two, two compute nodes against each other, but you can expand that to as many nodes as you want.
00:45:10
cyberpunkmetalhead
Oh interesting.
00:45:13
Blane Sims
And there's, you know, the technical paper talks about ah kind of the the benefits as you increase the number of nodes up to a certain point of you know increased kind of protection against collusion and other risks that you might be worried about. um But all of those nodes are going to compute in parallel. So you get an answer back, again, essentially and you know within a matter of you know sub second is kind of our, our, our target. And for, you know, for simple transactions, that's certainly the case. So all of these nodes are, are, uh, you know, sending their answer back. Um, the answer gets checked. Um, if everything, uh, lines up.
00:45:53
Blane Sims
Um, then, um, you know, essentially there's a first signal that says, you know, all of the nodes, uh, you know, had the same answer. Um, so, uh, there's a good chance that this answer is probably correct.
00:46:06
Blane Sims
Uh, and, uh, you know, think of that as kind of of a first level commit, uh, of the, uh, the compute answer.
00:46:06
cyberpunkmetalhead
Right.
00:46:13
Blane Sims
And then there's a, um, there's a limited timeline then for essentially objections to come in, uh, to say, okay, um, you know, maybe. one of the nodes didn't respond within the within the timeout. Maybe they they were saying, you're trying to censor me. They have a way to object that's outside of the ah kind of the core transaction stream so that they can that objection can't itself be censored. ah So there's a kind of a second stage commit that, you know again, depending on the sensitivity of what you're doing in your transaction, if you're moving millions of dollars, you might want to wait for that second.
00:46:49
Blane Sims
commit stage before you kind of stamp things as this being completely done. So all of those mechanisms seem a little similar to various types of of blockchain consensus that you might have seen before and that's you know very much kind of on purpose. ah But the difference is you know this result, yes, you can you can then take the output of that and use that as the basis for an input to a blockchain transaction or you you have no requirement to do that.
00:47:18
Blane Sims
So you can use this as part of the data pipeline that you're building an AI model on top of, right, so that you're, as you're pre-processing data, um you know, you have all of these source, we're working with a carbon credit project right now where
00:47:23
cyberpunkmetalhead
Right.
00:47:33
Blane Sims
ah You know, a lot of decisions are made based on satellite and environmental models that are, um you know, where data is coming in from various sources and it's not clear where um the data gets changed or could be changed that might impact someone's pricing of a carbon credit. So they're, you know, what they're doing, it's something called EcoMonitor.
00:47:53
Blane Sims
um They're creating a verified data pipeline that's showing from the source data. These are the transformations into the um you know into this state that then went through you know an AI model. you know That's a you know completely off-chain application where verification still is extremely important ah because if the data is modified or the model is um you know is corrupted accidentally on purpose, the output of that could swing the price of a of a carbon credit you know one way or another.
00:48:23
cyberpunkmetalhead
that's that's a very That's a very interesting use case and also I feel like a very important distinction as well. you It's ah it's essentially a decentralized network that could operate completely independently from the Web3 layer if if you wanted to.
AI's Role in Creativity and Industry Innovations
00:48:37
cyberpunkmetalhead
um
00:48:37
Blane Sims
Absolutely. Absolutely. Yeah.
00:48:39
cyberpunkmetalhead
that that That's super cool. and And also I think an important aspect that you've mentioned regarding AI, because obviously AI is very extremely relevant at this moment. And I feel like there iss ah there's going to be more and more discussion about the the ethics of AI. We've already made a start on that.
00:48:57
cyberpunkmetalhead
um And I feel like a a decentralized platform that is provably you know transparent could help reduce or or minimize the the ethical implications, right the kind of data you're using.
00:49:12
cyberpunkmetalhead
the way you're using it and and the way you create or reduce it the AI bias. like These are, I feel, the top three issues with the current solutions, the current closed source solutions, right chat GPT.
00:49:25
cyberpunkmetalhead
We know that the data is being hovered from the internet, right from from nonpro like nonprofit foundations like Wikipedia and like stock images and and all that stuff.
00:49:30
Blane Sims
Right.
00:49:37
cyberpunkmetalhead
um And we also know that there is bias coming from the training process where you know there's usually a human involved and that human then tells the AI, ah no, that's their own answer.
00:49:49
cyberpunkmetalhead
so You need to answer this way or that way and then that will go back and then they'll just give you these very you know weird responses as a result of that.
00:49:51
Blane Sims
Right.
00:49:58
cyberpunkmetalhead
So what I'm trying to say is that I feel like using a technology like yours to prove that look, our AI solution is it's fair and is transparent. I feel like that would be such a massive win for everyone potentially using AI in the future.
00:50:14
Blane Sims
Thank you. we We see the same opportunity. I think oh you know the the question is, just as we've seen in the you world of of blockchain and crypto finance, do we have to see the consequences of not having uh, the safety, not having the verification there before, you know, we, we put those safeguards in. Um, and that's, you know, I think there's a lot of discussion right now, uh, around how to make AI safe, but we haven't seen what happens when it's not, uh, to, to a great enough extent, I think for people to slow down, which is what, you know, what it takes. Like, can you show when you're, you know, chat GPT, can you show in your model where all of your data, where all your data came from? Uh, did you have the right to use that data? Um, you know,
00:50:59
Blane Sims
I think, you know, the the era of kind of just the creators of these these these giant models, you know, breaking things, taking no prisoners, etc., is still going on, and we're going to have to come back. and and and retrofit a lot of things, to you know whether it's the you know the the intellectual property rights or it's the you know the proof of something being authentic and real and not generated by AI.
00:51:24
Blane Sims
All of these questions, again, point back to you know how do you verify let's what's being stated as being yeah provably true.
00:51:25
cyberpunkmetalhead
here.
00:51:34
cyberpunkmetalhead
oh Yeah, exactly. exactly yes And not to mention the bias within the training process that can have all sorts of implications from you know all the way to to swaying like campaign results.
00:51:41
Blane Sims
Right.
00:51:45
cyberpunkmetalhead
And I feel like that is that is that is one thing that ah that maybe you know maybe an AI shouldn't do, is just not not not have any sort of political bias.
00:51:55
cyberpunkmetalhead
And I feel that is that is really important for for any any tool, actually. So when you have companies that have ah ah a political bias, training a tool with political bias to have this this bias you know within daily use case, I feel it's like people are not even aware um up of what of the reality that they're getting.
00:52:17
cyberpunkmetalhead
right If your search engine is biased, and if the AI you're using is biased, and if the ads you're getting are biased, then what is reality then at that point? So it's going a bit off topic, but I feel like it's a relevant thing to
00:52:27
Blane Sims
there Yeah, yeah, fair, fair, fair. And yeah, what, and, you know, all of the, you know, the, the misinformation that is presented, you know, with the, with an agenda on, you know, in any direction, right?
00:52:40
cyberpunkmetalhead
Yeah,
00:52:45
cyberpunkmetalhead
yeah
00:52:45
Blane Sims
What's the, you know, how do you get to the bottom of that, that problem and, and again, have at least a rational basis for, you know, for what is provably true.
00:52:57
cyberpunkmetalhead
For sure. I think we live in really exciting times, but also at times that are... Arguably one of the most confusing times as well.
00:53:08
cyberpunkmetalhead
People are are are a lot more likely to be to be fooled by but bye bye well buy everything but including AI.
00:53:08
Blane Sims
Yeah.
00:53:17
cyberpunkmetalhead
like Some of the AI videos that I've been seeing lately on on the internet, they're just there's they're so good. like There's no way that we're not going to have AI cinema in a few years from now.
00:53:28
cyberpunkmetalhead
They're just too good. every Everyone's going to have NVIDIA GPUs and NVIDIA is probably going to be like a million dollars a share by that point. um But it does definitely seem that it's getting to a point where it could be used professionally in in in the within the movie industry.
00:53:46
Blane Sims
i yeah I am always amazed you know almost weekly by some of the things that that are starting to be created for sure.
00:53:55
cyberpunkmetalhead
Yeah, it's wild. um So you mentioned that you had like that that one a really interesting use case, but I think that's one of the cool things when you're in closed beta um and when you work with a product that kind of offers a wide range of use cases for um other clients that you're in a unique position to kind of be witness to some of the cool things that people are building with it.
00:54:17
cyberpunkmetalhead
Are there any other interesting closed beta projects that you guys have observed that you'd be willing to to talk about?
00:54:17
Blane Sims
in
00:54:23
Blane Sims
Um, yeah, I think, uh, yeah, exactly.
00:54:24
cyberpunkmetalhead
that are obviously using Truebit.
00:54:28
Blane Sims
Um, well, you know, what is really cool about, uh, about TrueBit is the, yeah, the, kind of the applicability that, you know, we we talked about carbon credits. um you We see projects in the supply chain space, you know, project that we like talking about a lot because it's actually in production now, you know, even against our are kind of a closed beta system, you know, it's production quality system.
00:54:55
Blane Sims
So that that project's the TRIC project, which has been developed by our, you know, kind of a long time kind of partners of ours called Quadrons, who have the their own their own kind of supply chain focused blockchain.
00:55:08
cyberpunkmetalhead
Mm.
00:55:11
Blane Sims
um It's focused on, TRIC's focused on ah creating a circular economy ah within the textile and food industries, particularly in Europe.
00:55:22
Blane Sims
The problem that you know that quadrants faced as they were helping build trick was it's and essentially an integration, the the trick projects and integration of some existing ah you know on-chain and off-chain logistics kind of data sources.
00:55:42
Blane Sims
So they needed a ah mechanism to blend that data and particularly moving data
00:55:43
cyberpunkmetalhead
Right.
00:55:47
Blane Sims
from the hyperledger blockchain into ah you know into the quadrants blockchain and and doing it provably ah so that you know there was no doubt that the data was authentic. you know The signatures of the data were checked before it was moved and that it was written.
00:56:03
Blane Sims
exactly as it was read, um you know, was kind of one of the very first things that, you know, that we implemented with, which we would verify. um We've seen, so, IKU monitors one project in the kind of the data pipelines and an AI space. We've seen another project that's in doing modeling of on-chain data.
00:56:27
Blane Sims
So they're taking wallet IDs, creating fraud models against kind of wallet IDs and transactions. And their use of Truebit was on essentially using us as the verifiable inference layer to show that that the inference ah that was providing you know kind of a fraud score, if you will, for um
00:56:33
cyberpunkmetalhead
Oh wow.
00:56:49
Blane Sims
you know for the against the the model for particular wallet ID um was in fact you know referencing their their model and that wasn't you know just being completely made up. it's Again, the answer to that could could impact your ability to transact. I think some other interesting cases we've seen.
00:57:06
Blane Sims
So in some cases, interestingly, the enterprise around, you know, creating a, you know, a provable on-chain attestation for off-chain documents in the pharmaceutical space.
00:57:17
Blane Sims
That's super interesting. Retroubits playing the the role of the, essentially the notary that's certifying the attestation of a document.
00:57:25
cyberpunkmetalhead
Right.
00:57:26
Blane Sims
um And then a couple of kind of earlier, early on use cases we've seen are, In the kind of financial applications world, we've been working with ah a project that's creating a you know a new stablecoin-based remittance app, but there's some interesting kind of offline features of that where they they need to hold kind of cash and offline escrow and need a way to you know to safely kind of lock or unlock the offline escrow of cash as it's moving through their remittance app, and using Trubit as the backing for that.
00:58:03
Blane Sims
ah you know for that Iscope transaction. So it's really kind of you know kind of runs the gamut of both technology ah layers that we're sitting at, but also kind of the kind of the different industry problems that people are trying to solve.
00:58:19
cyberpunkmetalhead
that so That's really fascinating. There are some there's some great innovations um just happening. It's just fantastic. um It's really cool to see how many more use cases like a new technology unlocks. And it's always exciting to to ah just listen to that. Are people still able to apply to to become part of the closed beta?
00:58:37
Blane Sims
Yes, um so we're still kind of taking in a few final applications for our last cohort in our early access program, which is our closed beta. And then um again, as that out kind of gets underway in a few weeks, you know then that'll be the last round before we open up for open beta.
00:58:59
cyberpunkmetalhead
Awesome. And do you want to share the links if someone wants to apply to be participating in the early access? Where would they go?
00:59:04
Blane Sims
Absolutely, yeah, we'll we'll share links and please don't hesitate a date to reach out even if you just want more information, we'd love to talk to you.
Anticipation for Truebit's Launch
00:59:13
cyberpunkmetalhead
Awesome. Well, thanks a lot. um I think this has been fantastic. And I wish you guys all the best. I'll keep an eye out for your launch. I'm sure it will be um incredible.
00:59:22
Blane Sims
Fantastic, thank you so much.
00:59:24
cyberpunkmetalhead
Have a good rest of your day. Bye, everybody.
00:59:25
Blane Sims
All right, bye.
00:59:27
cyberpunkmetalhead
Bye.