Sep 29, 2023

Chainlink co-founder Sergey Nazarov spoke to CoinDesk managing editor Aoyon Ashraf during the Sibos 2023 conference in Toronto, Canada. The wide-ranging interview covers Nazarov’s thoughts on cross-chain interoperability protocol (CCIP), TradFi partnerships and tokenizing real world assets. Travis Detweiler from OPUS filmed this interview.

Video transcript

I'm a managing editor of Coindesk. I'm here with chain links, Sergey Nazarov. Today we're going to talk about what chain link is, why it matters and what chain link is doing with the tray and tokenization and what's in the future for chain link going forward, Sergey? Thank you so much for making the time. Let's start with it. Give us a brief overview of what is chain link. Sure, my pleasure. So chain link is the system that originated the centralized oracle networks and the centralized oracle networks are a computing environment that can come to consensus about everything that blockchains don't come to consensus on. So blockchains come to consensus about private key signatures, token ledgers and state machines. And that's the universe of things that they can come to agreement and consensus on oracle networks come to consensus on everything else. Market, data proof of reserves, random numbers for games, automation, more advanced computation. And now most recently cross chain and so the chain link system in total has processed over $8.5 trillion in transaction value using these oracle networks to basically process various data that triggers and manages those transactions. And while the chain link network began with market data, it expanded very far beyond that to be the largest provider of proof of reserve identity data, very widely used source of random numbers used by many of the top gaming companies and NT creators and now cross chain. And all of that is underpinned by the same security model. And that security model really focuses on generating validation and consensus on all these other things without a Blockchain, right? So you're generating consensus on these computations around data aggregation for market prices, proof of reserves, identity, data, random number generation, and now cross chain bridging, which is very important for both the public chain world and the bank world. So explain why that is significant for a crypto ecosystem. Sure. So I mean there are so many different, so many different projects, so many blockchains that are out there. Why is the oracle network like yourself is so important? Sure. So I'll give you an example. Centralized finance is considered the next iteration of Blockchain technology. But decentralized finance can't exist without oracles and oracle networks. For example, when we started launching chain link, the defi market was below 100 million. And then with the launch of chain link oracle networks, it grew to over 200 billion. Because what a DFI application is is actually that unchain state generated on a chain which holds the conditions of the financial product. But then it's also an oracle network that at the most basic level feeds, things like price data in order to settle and liquidate and validate the proper operation of that contract. Now DFI is actually a collection of oracle networks connected to a collection of unchain contracts because the D five contracts need automation. Many of them will soon need identity to take money from different groups and other things. Right? So now what DFI has become is a collection of smart contracts and the collection of oracle networks combined into one application just like web, two applications are a collection of different cloud services. So another way to think about it is that blockchains are kind of like an open database and all the APIS and services in the world are oracle networks. The other thing that I think oracle networks are going to provide now in a secure way is cross chain connectivity, both for messages and value. And the reason that's important is because the ability to move messages and value across chains basically creates an internet of contracts where people no longer need to choose a single chain, they can actually build their application across multiple chains. And even more importantly, the liquidity and the value in all the chains can be unified into a single network. And that can happen both in public chains and private bank chains and eventually public and private chains will combine to form a global internet of contracts and that we're using the world network obviously. So essentially what it is. If I was a and I wanted to put my bonds on a smart contract, I would need to get the price data and maybe need to. And then again, you have to say with different chains, but that's a long process. So what I would do, I would, if I were using chain link, then it, I don't even have to integrate with any other chains. I can just directly do my transactions and don't have to worry about all the other complex mechanism that goes behind it. A simple way of putting it. But I'm just saying, is that correct? That's largely correct. I think for trad fire, there's largely three stages. So stage number one is creating a tokenized asset on chain. Let's take something like a gold coin and through our proof of reserves system in chain link, we power the vast majority of the gold coins that prove anything about themselves. So you made a gold coin, but you need to prove that the gold is there. So you made a gold coin on your own chain. You've proven that the gold in the vault exists, maybe you've proven some other things about the quality of the gold vault provider or some other properties of the coin. But now you want that coin or that real world asset or in the trad five world, it would be called a digital asset to be purchasable or purchased by counter parties. Ok. Well, if those counter parties are not connected to your chain where you put out that Real World asset or digital asset, well, then your market is non existent, right? So your second problem is now I need to get connected to all these other chains, whether they're public chains or private chains where the liquidity can purchase my asset where the market is. Ok. So now you've made the asset you connected to the chains, someone else has purchased the real world asset, the digital asset from your chain. And it's been moved over CCP, the cross chain interoperability protocol, which is now getting very widely adopted both in the public chain world and also in the bank world, you've moved it to their chain, but now the asset is somewhere else and it still needs to be updated with the status of the gold. So now the proof of reserve system can flow that information and data over CP to the asset so that even if the asset goes to a third chain Blockchain c like let's say it's a chain of a custodian, then the asset continues to be updated. So it's not just about creating the asset and making it reliable and derisk by adding proof of reserves. It's also not about just moving it and selling it to another chain. It's about those two things and keeping the asset updated. And then the fourth equally important thing is that you should be able to do all of that without ever going to the chain where the asset ends up. So in today's world, you have this really weird experience for bridges where you have to take a wallet, you have to put the wallet, assets out of the wallet and into the bridge, you have to use the bridge, then you have to go make a wallet on the chain that you sent the value to. You have to take the assets out of the bridge. Put it in the second wall that you now have on the second chain. That's kind of a nightmare experience. What C allows you to do is it allows you to attach messages to the value so that you can just say from one place, you know, I want to send this stable coin over to this chain and I want it to go to this contract and I want to use this function. So let's go to Blockchain X to contract Y use function Z, for example, the buy function, I want to buy this much of a carbon credit asset and then send it back to me at this address. So in addition to all of these things of creating a reliable asset, moving the asset and keeping the asset updated. So it remains what's called a golden record or single source of truth. You can also do this in a very efficient way where you don't actually need to integrate or go to the second chain. You can just kind of say, hey, I want to send the stable coin here. I want to get the asset, I want to send it back. And this I think will greatly improve the efficiency and user experience and reduce the overhead that people need to use any chain because they can use it from where they are, they can stay there and they can use hundreds of different chains without going and integrating with, Of course, they can, they can send the value and the masses at the same time. So it makes it more efficient. Speaking of efficiency, um The the they need the efficiency to properly operate their business. So how are you helping? You had quite a few uh uh an from the world that you have partners with, partner with them. Can you tell us a little bit more about that? Sure. So I would say there's three big categories of things. The first one, the stuff that we've worked on with swift and continue to work on with them is the ability to connect to hundreds of chains from your existing infrastructure. So the thing that really separates banks from a lot of startups and even other mid size technology companies is that they were formed 40 50 years ago and they have all this existing infrastructure and all these existing systems that they're not going to get rid of. And they have all of these people trained on how to use those systems. And those systems are actually secure and used to hold value. It's actually from those systems that they're going to want to transact with blockchains. So the first thing was working with swift to make it so that swift messages, which is a very commonly used standard for banks to communicate with each other and internally among their own systems, for those swift messages to be used to trigger transactions on various chains. So now integrating into hundreds of chains is no longer about integrating into hundreds of chains. It's about integrating into chain link with one integration. And that one integration gives you access to those hundreds of chains. And this removes the friction and creates efficiency and makes it much easier for blockchains to become part of the infrastructure that banks use. Which means that the value that they have can flow onto chains much more easily. That's the first thing. The second thing is the ability to connect multiple different chains into an internet of contract into a network of chains so that the liquidity and the value in those chains can flow to any one financial product in that network. So that means you can launch a real world asset token on Blockchain A and then Chainz that's also on C I can purchase it. And the more chains that are in this network, the more value, the more market size, the more liquidity there is. So that second one is a very important one because the banks are going to make these assets to make them widely and easily accessible. And then the final one is actually making better assets. So you can now make tokenized real world assets where you have the input of proof of reserves, you can attach identity from an identity oracle network, you can attach price from a price network. And all of that creates a golden record or what's often known as a single source of truth that's updated second by second. And so anybody who owns that asset knows exactly what's going on with their single fractional share of value. They don't have to go and ask somebody, they don't have to go and figure out they know at every moment that there's gold backing this asset and the gold is worth this much and there's identity data attached to the asset. So it's legally ok for them to get that asset and this kind of de risking of assets by adding more and more information into them is really what the global financial industry and their infrastructure has been missing. And that's what I hear consistently. So on the one hand, there's all this opportunity in terms of connecting to other chains and getting assets to be used by other banks and purchased by them. And then on the other hand, you have all of these efficiencies that massively reduce the costs and very importantly, reduce the risk because the goal of the people here is yes, it's to open up new markets and have more opportunities, but it's to do it in a risk adjusted way that their risk is managed down because their favorite thing is not just crazy returns, it's returns that are derisk. That's actually the business that these banks are in. Right. That makes sense. And speaking of that risk, can you talk a little bit about how the risk management works for CC IP? And because obviously any kind of risk in this kind of in this world is, is is not acceptable for. So can you talk about how big of a important part of the CC IP, the risk management part of it? Sure. So CP took us over three years to build has undergone multiple security audits. And what it really is is multiple oracle networks working together to create a secure and reliable bridge where those oracle networks are checking key aspects of the security and reliability and risk management. So really every CP bridge can be divided into two key sections. One is the core transactional bridge, which is what's known in infrastructure as a thin pipe. So it's a very efficient, high throughput place where the transactions can move and they can move at high speed and very efficiently. Then you will have the risk management network which is a much more advanced network that approves or denies transactions without ever actually touching the tokens. And that network is where you encode all kinds of risk policies and risk conditions and so on. If you look at every system in the world that transmits value, credit cards, payment networks, bank systems, they all have risk management departments, risk management systems, risk management software. And that's because if you don't have a way to manage the risk of transfers, you will end up either taking risk injecting very large cost or having extremely low speed. So the dynamic with value transfer across the internet has been the people who can manage risk well, can achieve higher speed at lower cost with less risk. And so C is the only bridging system that has this notion of risk built in, in a way that is configurable and that you can add different parameters of risk. And that's actually a lot of what banks do is they have all kinds of systems that they analyze outcome, but they also manage the risk. So you're saying that the the parameters for the risk management is customizable, completely configurable. In fact, the whole chain link framework is made to make completely configurable oracle networks composed of whatever collection of nodes the developer feels are best for them and their security needs with whatever in the case of CP with whatever risk policy and risk management practices that they feel are the best to guarantee the security of their transactions. And the reason that's very important is that a we don't know all the risks that are yet to unfold in relation to for a value across chains. And so it's very good to have a flexible system that can incorporate those risks as they appear, which is by the way, how all value transfer systems work and manage risk, they always have very robust ways to add more risk management. Eventually, I think the risk management network will even incorporate elements of A I and all kinds of advanced analysis. And then the second thing that I think is also important is the ability for the system to be easily and quickly integrated into all the places where the people using it want to get transactions from. And so CCIS I think on the bank level is very far ahead on that side, based on all the banks we're working with as well as the cds like DTCC, euroclear and various others that we're in conversations with. So I think we are kind of in a place where you have these two parallel worlds, you have the public chain world and then you have the bank chain world. And while these worlds are kind of separated by a lack of legal clarity, once they both run on C IP and once that legal clarity appears and that wall goes down, they will all transact with each other on the basis of using the same connectivity layer and on the basis of the ability to define their risk as they transact with each other. Fascinating, let's switch gear a little bit let's talk about your competitors and your market shares. So in terms of TV S total value secured and on an enterprise basis, you guys have the largest market shares. Why is that? And what are you doing to stay ahead of your competition? I think the fundamental answer is security. So if you look at the competing systems, many of them are really quite centralized running on one or two servers. And then there are other systems that have something like 20 servers. But then it turns out one person has the keys like multi chain is an example. So I think the centralization and the lack of real reliability and security and the way that these systems are architected, it just makes it obvious that they are not made to be reliable or secure in many cases because the people architect them never built a system like this. We have a very large community of security experts and researchers at chain link labs, there are hundreds of people working on helping improve the security of the system as well. We have top academics and researchers working constantly on verifying the security and there's millions of dollars spent on the security audits of every aspect of the chain link system, whether that's the data side or whether it's the cross chain side. And so I think the fundamental value of security in our industry is never to be underestimated. And chain link as a framework successfully uses decentralization to create real security, which is proven by the fact that it's securely and reliably processed over $8.5 trillion worth of transaction value. What time frame that we only started counting? That's only counting from the beginning of 2022. So that's for less than two years. But the system has been live on production without any significant issues for over four years. And the second reason is reliability. So the first one is security and immune to the immunity of loss. And then the second one is reliability. So for example, when Ethereum Gas went to over 2500 many systems stopped working, the majority of systems stop working. But chain link networks continue to update without missing updates. And so on the one hand, it's security. On the other hand, it's liveness and reliability. And then the third side of the equation I would say is really the quality of the connection. What are you connected to? So the data quality of the chain link network really can't be surpassed in terms of any of the data that you can get for identity, price proof of reserves or anything else. So the actual inputs into the chain link system are the highest cost, highest quality inputs that you can get right now for defying all these systems. Many of the other ones are easy to manipulate. They're cheap or they're actually built in a way that they're meant to manipulate the application that uses them to extract value from that application, which is a very weird predatory model. I'm starting to see emerge. And then on the cross chain side, the ability to securely and reliably connect to the right places is something that chain link and CCP are continuing to get more and more of. And I think on the bank side, we'll have by far more than anyone else. And we're doing that once again in a way that maintains the security and reliability of the system. How are you doing that? So it's really how the protocol is built. So I'll give you an example. So on the data side, when FDX happened, there was a multitude of data providers that were heavily weighted towards FX. And the chain link system removed the data providers and replace them with reliable data providers that weren't heavily weighted toward FX. And that's an example of risk management. And that's something that the system kind of did and was architected to do. And that's something that, you know, the people running the system, the nos that everyone was involved in. On the cross chain side, we have the risk management network and no one else has any real notion or module for risk management, the whole bridging system, which should kind of tell you how little real principled security design has gone into those things. So I think you can't really add security later. You have to do what's called principled systems. Design. And for example, another consequence of our principal system design, flash loan attacks. So the gene link system on the data side completely avoided using TP. And basically the dex prices that are very easy to manipulate. And there were a number of people that didn't necessarily believe us. And we're talking about how T APP is great and it's fine and then guess what? Those people got a flash loan attacked because that's not a principled systems design decision. And the reason that chain link has resisted all of these flash loan attacks is because it was designed in a way to resist them, not when they started happening, but before they started happening. So it's not about adding security later. It's about making sure that security is built in from the beginning and then going through a rigorous review process and auditing process with multiple auditors and actually spending millions of dollars on that, which many people are unwilling or unable to do being proactive is always better than being reactive. Get more of a broader picture. What's your view on the tokenization? The real? There has been a big push from the fight the banks coming in and that token their real world assets. What's your view on that? Do you think this is the future? I think, I think it's going to be a very active market for banks. I think they're good at doing this because they have access to all the world's real world assets and real world asset holders come to them to turn them into financial products already. So I think they're very well positioned to generate many real world assets and I think they see a big market for that. I'm seeing a lot of stuff in carbon credits, some fund organizations, some real estate and other things. I think initially they were going to transact with each other because they know each other as counter parties and they feel comfortable legally doing that. I personally believe that eventually the biggest market for real world assets from banks will actually be public Blockchain protocols because the public Blockchain protocols need diversified collateral. This is something that they desperately need to avoid ending up like terror basically because they don't want to be heavily overweighted towards crypto only collateral because when there's the cyclical bust of crypto, the cyclical bust results in orders of magnitude greater problems than if you were diversified, even to 30% of your protocols collateral. And they, I think the public Blockchain protocols are the ones that will be willing to pay the biggest premium for this diversified collateral. There is just once again, this wall, this legal wall between the public chains and the private bank chain world. We are supporting both of those worlds through proof of reserves, data, identity, various types of computations and cross chain being built out in each of those separate ecosystems. And then I do very strongly believe that the yield from the public Blockchain world will be very attractive to banks and they will put their clients' money into the yield generated by public chains. And then the public chains will greatly benefit from the assets that the banks tokenize and put into their protocols, making those protocols more resilient and reliable. So I think it's the relationship doesn't exist because people don't see it happening. But to me it's very obvious that the economic interests of these groups and their financial benefit to each other can be aligned and we're building the technology and the rails and the systems that will allow them to align in a very informed and risk managed way so that everyone can look at the data backing the assets, everyone can move the assets and even as the assets move, they will continue to remain updated and reliable. It makes sense in terms of all the conversations that you're having with these banks. Is this somewhere somewhere where they're heading towards? I mean, I don't think they're heading towards. I think some of them are there like we announced some stuff with Anz one of the largest banks in Australia with over a trillion dollars in assets under management where basically using c there was a transfer of a stable coin from one chain, their stable coin that they made from one chain to another chain in exchange for a Reef credit, which is a type of green asset, carbon credit type of asset that was then moved back to the originating chain that paid for it with a stable coin. And that's something we were able to make work with the help of CCP and various other technologies from the chain link framework. So I thought, yeah, that was on test net, but I'm seeing these things move forward. So I've never seen a few things. I've never seen people with digital assets in their title. So there's entire teams of people now in the banking world that are focused on digital assets as their full time and only job. It's not in like the innovation group. And now I am seeing them go to production and be able to conduct transactions with each other also because there's a little bit of more legal clarity like in Europe, for example, there's the pilot regime for tokenization that came out in March and then in other places, there's some amount of legal clarity that allows them to transact on a secondary market basis. And so I am seeing them go to production and I think we're going to be pretty involved in that. It makes sense. You touched a little bit about the regulatory clarity given what's going on in the world right now with what's going to u what's your view going forward? How does the crypto world move forward? I think there's a fundamental truth about the crypto world that it's good for society. And I think all of these governmental systems and frameworks. They want to protect society from bad things and they want to encourage good things. So I think the truth of the matter and the truth of the situation is on the side of blockchains, similarly to how it was on the side of the internet. And when the internet appeared, there was a lot of weird activity on the internet. People did weird things. There were new situations that the technology allowed to happen that previously were not, didn't exist and it was kind of like a little freaky. But what the US did and what other people did is they were permissive about what you could do on the internet and they allowed people to transact and they allowed people to build valuable applications. And now we're all on the internet, right? People will be watching this over the internet, we communicate over the internet, we consume media and content from the internet, the internet gives us transportation through things like Uber. So I think at the end of the day, the net benefit of blockchains is so massive that even if there are some negative consequences, just like there are with the internet that all of these governmental structures, regulatory structures will see the value of blockchains and they will implement them and guide them into a way that's useful for society. So I think the first thing is that the truth and the value of the system is very much the truth of the value is there. It's absolutely there. There's no doubt about that. And so it's all going to end up fine for that simple, basic reason that this is beneficial to society. There will be ups and downs, you know, the industry is very cyclical. I've been in the Blockchain industry since 2010 back when it was coined, called the Bitcoin industry because bit one was the only Blockchain at that time And I've seen it go through many cycles and the cycles are always the same, you know, everything booms, people make money, everything grows. And then the system can't handle that much money, can't handle that much value. And then somehow it breaks and then someone loses money and then cyclically things go down. But every time the boom comes back, right? Because the value of the technology is there and what it allows. And so I think now we're just in a place where we kind of reached the pinnacle and the height of the industry that wasn't previously reached and things didn't work out, you know, a bunch of things broke. We learned a bunch of ways that things shouldn't be working. Now, those things are hard lessons, everyone's learned, we fixed the system or the system is in the process of getting fixed. And so next time I'll be able to handle more value. But that's what I think all of this stuff is a reaction to is something didn't go right. Something didn't go. Right. Doesn't mean that the fundamental value of something isn't there when you make new things where you push the limits of what's possible, when you invent new ways of doing things, you naturally discover ways that they're not supposed to work. And, you know, that's just not something that everyone is always used to. I'm used to it because, you know, in building new technologies you see on a common basis on various technologies. But I think it just takes people a little bit longer to kind of step away from whatever happened, that was wrong and to get back to realizing the fundamental value of what's going on. But the very important thing is that the truth is on our side because the value is there and it's there for the individual, it's there for the government, it's there for society, it's there for the banking industry, it's there for everybody. So I think at the end of the day, that is what has always driven the very large adoption of every technology, even if initially there were all kinds of risks and problems. For example, with cars, the early days of cars, there were laws that said that before a car was able to enter the city limits, someone had to walk in front of the car and wave two red flags and announced to the town that a car was entering the town limits because the car was classified as an explosive device because it, it was using explosions to power the engine, right? Or was using explosive fuel and now we all drive around in cars. Right. So I think it's just that recurring pattern with technology that we're seeing a version of now makes sense. If you were to compare the, the early days of internet versus crypto, where would you say? We are? I can't tell if we're pre 2000 or post 2000. I can't tell if the hype cycle has reached a kind of peak or not. I don't think it has because the vast majority of the world's value has still not gone on to a Blockchain. So I would say that we're pre 2000. In that analogy, I also think from an infrastructure level, a lot of the infrastructure isn't fully built out or even at a basic level. And that infrastructure is also, I would say in the pre 2000 stage, I think the infrastructure will grow. I think it will be able to handle more value, more speed, the risk will be managed better with the help of proof of reserve and other technologies. Then I think the industry could easily boom well past one trillion into the double digit trillions from the 10, 20 or 30 or more trillion dollars total market cap range. And then we will see if the system can reliably manage that. We can see if blockchains can reliably hold and secure and be responsible for that much value. And if they can, then the system will continue to grow. And if they can't, then we'll have another cyclical cyclical boom and bust cycle. But I think we're still in the internet analogy, we're still in the pre 2000 stage of this industry fair. So what's next for chain link? I think what's next from the trajectory that we're on is continuing to process trillions of dollars in transactional value. Injecting the most amount of reliable, validated data into both D five public chain applications, Blockchain games, insurance NTS. And now the banking industry, I think a big focus is going to be on cross chain and C and creating this internet of contracts and public chains and this internet of contracts and private bank chains in anticipation of both of those internet of contracts merging into one big global internet of contract. Similarly to how TPI created the internet, we're working pretty much day and night to see how C I can generate a global internet of contracts where all the chains are connected. And then once you have a way to inject all the information, do all the compute and provide all the value across chains in a seamless efficient way. I think you get to the next level of applications. And similarly to how Uber and Netflix and those applications define the internet for most people because the infrastructure got to the level that it could enable those applications to exist. I think we are now kind of in the stage where the infrastructure that made something like Netflix and Uber possible is being built right now by us. And then as it gets built and as it reaches those scales and scalability and security needs of that quality of application, we will see another crop of applications that are, you know, at the level of that they're like the Uber or Netflix equivalent of Blockchain industry. And then those applications will drive the mainstream adoption of the industry, both from the retail, basic consumer side and from the private bank side. And then eventually both of those worlds will once again merge to become a single global internet of contract, right? Makes sense. And speaking of the banks, do you have any partnerships that are you working on that you can share with us? I have more partnerships than I can count at this point. We're actually going to after this conference here, we're going to take a tally of the amount of people that we're going to be working with on various different pcs and pilots and going towards production with some people. I think the amount of demand is quite large because banks, they have the resources, they have the team members and they have the market that wants these real world assets from them. And now the question is really twofold. How do you technically enable all that to happen through CCI and proof of reserves and identity data and all these other kinds of key inputs. And the second question is how does the legal framework that banks are so dependent on allow and enable and accelerate that? And what what we're going to be doing is getting set up in a technical sense to help facilitate all those transactions. And then as the legal clarity comes the amount of adoption, I think will skyrocket because that's the only thing that I really see holding banks back. It's not the market opportunity. It's the lack of clarity on how to scale the market opportunity into the size that they want. And it's partly the lack of technical rails and systems that can do that efficiently. So we're going to solve the technical rails efficiency problem and then the legal problem will gradually get solved and then we'll kind of hit an inflection point where the amount of throughput I think will go through the roof, in my opinion.

Learn more about Consensus 2024, CoinDesk’s longest-running and most influential event that brings together all sides of crypto, blockchain and Web3. Head to to register and buy your pass now.