Oracles and the Expansion of Blockchain Utility (w/ Sergey Nazarov and Ash Bennington)

ASH BENNINGTON: Sergey, welcome to Real Vision. SERGEY NAZAROV: Great. Great to be here. Thank you for having me. ASH BENNINGTON: It's a pleasure to have you
here. This is an interview that I've wanted to do
for some time now. We first met, I guess, a couple of years ago
at a conference in Brooklyn. We had a conversation where we talked about
what you were doing, and I was just spellbound by it.

And it's really a pleasure to have you here,
and to get you to have the opportunity to explain a little bit about what you do. So let's just come jump in and get started. Starting at the very beginning, what are smart
contracts, and why are they so significant to the global financial system? SERGEY NAZAROV:
So smart contracts, in the simplest terms, from a user's point of view, are math-based
contractual agreements.

And I think the way to contrast them is to
think about brandbased contractual agreements. And this will highlight, for people, the notion
of something called "tamper-proof-ness," and the usefulness of "tamper-proof-ness," because
a lot of the times, blockchains are explained in these technical parameters of "tamper-proof-ness"
and reliability and censorship resistance and immutability, and these concepts don't
always transfer over to use cases. So the thing that I think is needed is an
understanding of a contrast between how the world works today, and how the world will
work in a math-based agreement kind of world.

So the thing that people maybe don't think
about so much, because it seems to be working, or it works in certain cases, and so is good
enough, is that the agreements they have around their bank accounts, their assets, financial
instruments, are really brand-based agreements. And what brand-based agreements are are a
logo. There's a logo on a building somewhere, just
like in the Wild West. The biggest building was the bank with the
pillars, and everybody would put their money in the bank with the pillars, because it would
have the really nice pillars, and be really impressive, and it would be the nicest building–
nicer than the church. And that was the brand-based guarantee that
that bank gave you in that kind of gold mining town. And things haven't really changed very much. In reality, over– I would even say hundreds,
maybe even thousands of years, where you basically have a brand-based guarantee. I have a logo. My logo represents an institution or an entity.

My institution or entity has been around for
x 100 years, and I guarantee to you, on paper– on a paper document– that my relationship
with you is such that your assets or your financial product or the value you hold in
my institution is within your control. You will be able to access it. You will be able to buy and sell financial
products. You will have a savings account that you will
always be able to access. All the money in your savings account you
can always access. All the money in a trading account you can
always trade, and you can always liquidate it whenever you want, and all these types
of guarantees are foundationally based on a logo.

They're based on I exist for a long time. I will continue to exist. I'm guaranteeing to you that this relationship
that I manage for your value that you hold with me will function in a certain way. And so those are brand-based agreements. Now, what math-based agreements do is they
basically say, there's no brand. There's no thing that has existed for hundreds
of years, and I promise you, things will work a certain way in a paper document. There is just math. So there's cryptography and mathematics that
guarantees, at the level of physics and mathematics, that you have a certain relationship with
a certain asset, certain financial product, certain contract. If we begin to unpack this and we think about,
what does that mean? Let's take Bitcoin as an example.

The relationship that a user has to a Bitcoin,
regardless of the value of a Bitcoin, regardless of the value– RAOUL PAL: Hi, I’m Raoul
Pal. Sorry to interrupt your video – I know it’s
a pain in the ass, but look, I want to tell you something important because I can tell
that you really want to learn about what’s going in financial markets and understand
the global economy in these complicated times. That’s what we do at Real Vision. So this YouTube channel is a small fraction
of what we actually do.

You should really come over to realvision.com
and see the 20 or so videos a week that we produce of this kind of quality of content,
the deep analysis and understanding of the world around us. So, if you click on the link below or go to
realvision.com, it costs you $1. I don’t think you can afford to be without
it. SERGEY NAZAROV: this and we think about, what
does that mean? Let's take Bitcoin as an example. The relationship that a user has to a Bitcoin,
regardless of the value of a Bitcoin, regardless of the value that a Bitcoin attains within
the global market, is that they have something called a private key.

So they have a cryptographically enforced
way to sign and to mathematically prove that they have, and only they have control over
this digital good housed in this network of thousands of computers. And so I'll give you an example of where this
really starts to make sense. So if you look at the debt crisis that there
was in Greece a few years ago, you saw people only being able to withdraw 66 euros per entity
and per person from an ATM. So whether you had a business as an entity,
or you had an individual, they could only draw 66 euros. You have similar situations in certain countries
where things aren't going as well as they could be, and you have ATMs and bank controls
being locked up.

What you also see, in all those geographies,
is wallet registration numbers growing by 300%, 400%, 600%, because people basically
realized that their relationship with a certain asset was not what they thought. They thought their relationship was very deterministic. It was very direct. It was on the level of physics, and a certain
amount of surety and guarantees. And what the reality is– it's very different. It's like insurance policies. People don't know what's in their insurance
policy.

Realistically, people that rely on brand-based
agreements to underpin their relationship with their assets have that same level of
counterparty understanding. People are universally surprised when, oh,
something goes wrong in a country, and the ATMs get locked up, and they shift to Bitcoin. And the reason they shift to Bitcoin is it
doesn't matter if you have a billion dollars in Bitcoin. If you have that private key, if you have
the mathematical proof in your hand to move that asset, there is no way that that can't
happen short of physics and mathematics starting to work differently. And so this is the unique difference between
brand-based and math-based guarantees. And this difference actually goes much, much
further than people think, and part of the reason that this difference is underappreciated
is because– well, basically, while everything's working, everything's fine. That's basically how the global financial
system works. Everything's working, so everything's fine,
and you don't really need significant improvement for your assessment of counterparty risk,
for your management of counterparty risk, because, well, everything's working.

I can access my assets. I can liquidate them. I can do whatever I need to do with them. But let's say there's an unfortunate possibility
that the global financial system is now in a direction where its underlying solvency
and its underlying assumptions might result in certain issues where assets can't be accessed. Counterparty risk becomes apparent in how
faulty it is as a way to hold value, or hold assets.

And this is where the uniqueness of math-based
agreements, even in their simplest form– Bitcoin is like the simplest form of a math-based
agreement. It's the first one. It's the simplest one. It's the most secure one. It's the most adopted one. But it is, by no means, the only one. And there are many, many other forms of math-based
contractual agreements that have taken the form of other tokens. Now we're starting to take the form of DeFi–
decentralized financial products. But what they all have is they fundamentally
have this level of control, and this level of transparency and counterparty risk assessment
and mitigation, that traditional financial markets don't have. And I think there's a very important difference
there that's going to become more and more appreciated as that problem surfaces. And then as that transparency of these smart
contracts that work in these systems begins to get valued, you also see people composing
them in very useful ways.

So fundamentally, from a user's point of view,
it's a math-based contractual agreement. From a technological point of view, it's basically
a piece of code that represents an asset, or represents ownership, or represents a financial
product, but this piece of code operates in a system of computers beyond the control of
any one party. And it's beyond the control of the person
holding the value. It's beyond the control of somebody who'd
want them to stop from liquidating their value, or controlling their value. And that's the uniqueness of it. And then because you've now taken contractual
agreements, financial products, and ownership, and you've turned it into little pieces of
code, now you can have many, many smart people build all kinds of configurations of those
pieces of code, and that involves creating an ecosystem for that– which is partly what
we're involved in– and then once you create that ecosystem of people being able to combine
these little math-based contracts in the forms of pieces of code, you'll start to see an
explosion of new financial products, new assets, just like you saw an explosion of what internet
companies could do, because once internet companies could compose little individual
pieces of code into chains of code that interact with each other in meaningful, new ways, you
saw ecommerce, you saw Uber, you saw all these other kind of innovations that were really
the composition of existing other pieces of code together in unique ways.

And so you'll, on the one hand, get a system
that gives people an ability to properly manage counterparty risk. And on the other hand, you'll get an ecosystem
where people can finally compose and build financial products the way internet companies
build web products– at the same speed, and at the same usability that people expect from
the web. ASH BENNINGTON: Yeah. There's so much there, but two things really
have jumped out at me. The first is just how truly profound that
shift is between a brand-based relationship, as you call it– some others might call it
a relationship-based– and physics and mathematics on the other hand being the true underlying
bond between counterparties. And the second point that I thought was very
interesting is how truly globalized this effort is– and you touched on that in a number of
different ways. There are many people here who have been fortunate
enough to grow up in the United States whose parents and grandparents have been enmeshed
in this system for 75 years, and they've never had a significant incident.

They may have lost a job and had trouble paying
the mortgage, but no one ever questioned the title on their house, and their bank account
was always available. But that has not always been the case throughout
the rest of the world, and I think it's crucial for people to understand how this is truly
a global challenge, and also a global movement that is moving to solve it. Sergey, to follow up on some of the points
you made, you've just described the case for Bitcoin, the case for what the underlying
value of the technology is. But what specifically are smart contracts? What differentiates smart contracts from Bitcoin,
for example, and how do smart contracts continue to extend the existing functionality of an
asset like Bitcoin? SERGEY NAZAROV: Sure, sure. So smart contracts are basically code. They're business logic that define relationships
between assets, between data, and outcomes. So they're just code in business logic. But they're code and business logic that operate
in this unique, tamper-proof, math-based guarantee universe.

So it's basically the difference– it's almost
like the difference between– well, it's a difference I said, right? It's brand-based and paper-based I promise
I'll do something, and when a system says it's going to do something, physics forces
it to do it. And so it's basically code that operates in
this universe. What it really does is it extends– it's an
extension of what you can do with financial products in this universe. So if you have the math-based guarantee universe,
and before you only had let's generate a token– first, really, the evolution of this was first
Bitcoin. And you had one ledger for one asset on one
network of computers only focused on that one piece of data, and that network is only
interested in, hey, how do I maintain and secure this one ledger of ownership about
this one asset called a Bitcoin? And I have thousands of computers doing that,
and I have a lot of hash power and security assigned to that, and I'm securing this one
ledger.

So that was the beginning. And then you had an evolution where people
started to say, well, hey, I want to apply this universe– this math-based, contractual
guarantee capability– to other things. I want to tokenize art. I want to tokenize equity. I want to tokenize land. I want to tokenize whatever I want to tokenize. I want to create a token, and I want to put
it in this format, essentially for two reasons. The first reason is this math-based guarantee
reason, and the second reason is because due to the internet, when I tokenize something
in this format, this becomes globally accessible.

So I can make a token that people can interact
with globally, instead of I make a token, and I put it on some small platform, and only
that platform's universe can interact with it. And that's part of the reason why blockchains
and tokens have taken off, where basically, even if one geography at any one point says,
hey, I'm not into this. I'm not into this token thing. There's 15 other geographies that are into
it. And then if one of them says, I'm not into
it, well, they're not saying it at the same time.

So even if one geography says, I don't like
tokens, the other geographies are either behind or ahead on their thinking– depending on
how you look at it– and they're all saying, we're fine with this. And so there's still a global market for tokenization. But tokenization is not– it's a very simple
smart contract. And what it actually does is it basically
says, I'm going to generate a list of assets. I'm going to make a digital list of a million
things– a million units. And then I'm going to track where those units
go. I'm going to see the 10 units go to you, and
10 units go to you, and 100 units go over there, and I'm going to maintain a ledger
that basically tracks who has what units, and I'm going to maintain that history of
ownership.

And that history of ownership is really what
a token is. It's a token, the issue of ownership, and
then the token represents something. And that's the story that people craft around
what that token represents. Does it represent a house? Does it represent a piece of art? Does it represent a digital good of some kind,
or equity, or whatever it represents? And that is the next– that was the next evolution. And that evolution– we can thank Ethereum
for that evolution. Ethereum did a very impressive thing where
they took the space from, hey, you can only have one token, or it takes you years to make
a token, to hey, it'll take you a day to make a new token. And so that was the functionality that Ethereum
enabled. It enabled the functionality of, let's make
tokens. And so because that functionality finally
became polished to a certain level of usability while maintaining security– and that's actually
an important aside here, that the point at which things in our space begin to get adopted
and utilized is once they can be built by developers without the developers having to
build the security.

So once infrastructure creates enough security
for the developers to efficiently generate things, that's when things get built. So first you had Bitcoin. It generated security. It generated Bitcoin. Then you had Ethereum. It generated security around making tokens,
and it allowed people to generate tokens, so there was a big boom in token generation. And that was a smart contract, but a simpler
version. And now you're moving on to more advanced
smart contracts, where the business logic is basically saying the types of things that
financial products in the real world say– let's make a derivative. Let's make a futures contract. Let's hedge some kind of risk against a market
event, or a real-world event like weather. And these more advanced smart contracts, they
are really the thing that I've been working on for, I think, over seven years now.

And the thing that has got me into this space,
and got me very excited, because it applies this math-based guarantee to such a large
universe of contracts. Whether that's insurance for emerging markets,
so that farmers can have crop insurance and be able to withstand weather events, and that
that allows them to continue operating a farm, even though there's a drought or something. Whether it's financial products, or savings
accounts in emerging markets, where people wouldn't usually have a savings account to
combat inflation, or even whether it's more advanced contracts within developed markets,
where you're properly managing the risk around transparency, and mortgages, and all these
types of things. So, I think that smart contracts are this
tamper-proof, mathematically guaranteed contractual agreement. But they're mathematically guaranteed specifically
because they're run in these networks of computers that are run in a very specific way. And this is what decentralized infrastructure
means.

Decentralized infrastructure equals math-based
guarantees. Anything other than decentralized infrastructure
means brand-based guarantees. And that's, I think, the important mental
leap for people to make in their understanding of why does this matter. It's because– this is the problem, if I explained
the Internet to you by saying, hey, there's this cool thing called HTTPS, and it'll let
you move around credit card numbers. And once you can move around credit card numbers,
it's going to be great. You would sit there, and you would go, why
am I, what? HTTP what? Credit card numbers, why? You would just be completely like, why do
I care? But when you really think about it that HTTPS
has enabled people to move credit card numbers led to e-commerce.

ASH BENNINGTON: Right. SERGEY NAZAROV: And e-commerce led to all
of us being able to sit in our house, and click a button, and have things delivered
to our house. ASH BENNINGTON: Right. SERGEY NAZAROV: And so that's the connectivity
between new infrastructure and new use cases that is not as simple as like, oh, yeah, they
released a new version of a phone, and it makes a new interface where I can now order
things in my hand. Like that mental leap is shorter. It's like if people can order things from
their hand, they will. OK, that's a mental leap most people can make. To understand the history of infrastructure
in web-based, network-based systems and then extrapolate how that history evolves into
new use cases, that's not a natural leap. So, that's where I think there's this disconnect,
that once people get past that disconnect, and once they see math- based agreements and
their value, and that decentralized infrastructure creates that, they begin to see value in that. And smart contracts are just the name for
that. So, they are the name for– like if you had
a choice, do I want a brand-based agreement where the insurance company might pay me out
or might not pay me out, or do I want a smart contract that guarantees me payment if something
happens, every rational actor would choose the math-based smart contract.

And that's the guarantee that smart contracts
have, which other systems don't. ASH BENNINGTON: Yeah. I guess we could joke a bit about whether
humans in practice are, in fact, rational actors. But leaving that to the side for a moment,
that framework of understanding the way these networks work, it brings us right to the present,
which is the challenge of networks searching for a source of truth, the tautological nature
of a relationship within a network, and to precisely what you're doing at Chainlink now. Sergey, tell us a little bit about what the
background of the oracle problem is and why it's so significant in this space. SERGEY NAZAROV: Sure, sure. So, the oracle problem essentially is the
question of how do I on unblock access to all the world's data for these math-based
contracts? That's essentially the problem. And the problem exists because blockchains
only know about what goes on within them. Blockchains only generate their own data,
and security, and math-based guarantees about what goes on inside of them.

And the only things that really go on inside
of them are token movements, signatures from private keys, and a few other things. And that's a very important foundation that
we need to have for math-based agreements to even live somewhere. So, the math-based agreement and the logic
of the agreement can live in this network of computers that are purposefully limited
in their interaction with the outside world in order for the math-based guarantees to
be maintained, in order for the security of a smart contract to exist at all. But what that does is it basically says the
only thing I can write a smart contract about is the thing that I have information about.

And the only thing I can have information
about are things within the walled garden, within the blockchain itself. Which, to our earlier example, is tokens. Because blockchains need tokens in order to
pay miners. In order to pay the computer systems that
secure them, tokens as a functionality are inherently built in. And so, a token contract, it generates its
own data. It generates a lot of token units. And it sends those units to different addresses. And all the data is encapsulated within a
blockchain, which is why blockchains so far have been about tokens, even though math-based
contractual agreements and guarantees can be applied to so many other things. What an oracle does, is an oracle enables
the logic in a blockchain to interact with external proof.

So, it enables it to know that goods have
arrived, that a weather event occurred, that a market price has changed, that's some contractual
outcome somewhere has occurred. And I think the important nuance to understand
here is that you can only write automated financial agreements, you can only write automated
math-based contracts around things that you can prove. Because the math-based contracts are basically
saying, I have proof something happened, and so I'm going to act on it. And so, you could only really write a contract
in that format around the things that you can prove. And so, then the question becomes, OK, I want
this math-based contractual format. I want more contracts in that format. How do I enable– how do I enable that? How do I achieve proof to the level that it
can be considered math-based contractual agreement proof? That it can be considered so reliable such
that it could trigger a billion payment about a weather event, or it could trigger a $100
million financial product? And so, this is the nature of the oracle problem,
is firstly, how do I even provide this data? But then, secondly, the second order problem,
that we deal with, the harder version of the oracle problem is, how do you prove that something
somewhere happened in order to trigger something in this system? Because what would you should seek to achieve,
and we seek to achieve this, once again, through this decentralized infrastructure approach,
is you should seek to create as much proof as possible that an event occurred.

And the more proof you can create, the more
automated you can make your agreements around that event. Like, for example, if I have 100 sensors all
telling me that the weather, or the rainfall, or something is in a certain state, I have
a certain level of proof that I can pay out an insurance policy. ASH BENNINGTON: Right. SERGEY NAZAROV: And so now, I can automate
that pay out of that insurance policy. And if I can automate the pay out of the insurance
policy, well, then guess what, I don't need a local legal system to enable insurance. I can have a parallel technologically enforced
legal system that will provide crop insurance to people on their phones in all parts of
the world, where it might have taken them decades to get crop insurance against droughts,
and weather events, and all these things that impact all these people farming.

So, I think that the oracle is essentially
creating definitive truth about the external world. And it's pumping that into another system
that has definitive truth about a contract's state. And once you combine these two systems, you
open up a world of implementations and applications for these math-based contractual guarantees. ASH BENNINGTON: Right. SERGEY NAZAROV: Which change people's lives
in the emerging markets and substantially change the way the global financial system
works in the developed world. ASH BENNINGTON: Yeah. These are just fascinating questions. So many points, but the first thing that comes
to mind is the technical challenge of synthesizing all of this data.

To use a simplified metaphor of the point
that you were making, if it's a smart contract that is a claim due to weather for example,
if there's a billion at stake, one might plausibly assume that a bad actor could go and stand
over a sensor with a watering can and fake the data. But when you're drawing the ultimate source
of truth from a network of hundreds of sensors, you can build in ways of analyzing that data
to make sure that it's accurate and to provide true falsifiability. What are those challenges like? And how do you think about them at Chainlink? SERGEY NAZAROV: Yeah, it's a very complicated
and fascinating problem. We're very lucky to work with some of the
best people in the academic world and the security research community on this problem.

So, it's not– it's a novel and very important
problem. Because what you're essentially saying is,
how do I make computer systems that are provably independent from each other, and un-gamble,
and un– incorruptible? How do you get them to agree and prove that
something happened? And then how do you connect the world's data
into such a system to prove that things are happening a certain way? Whether those are claims about political events,
whether they are claims about whether, where their claims about price data. And so there's a few nuances there. One of the nuances is proving that data came
from a specific source, which is something that we work on very actively. And we have a lot of approaches, both cryptographic,
and kind of software based that we apply to that. Another dimension is guaranteeing that the
data will be delivered to the smart contract system. So, you need to prove that the data came from
a specific source. You need to have a capacity to include many
different sources, or as many sources as you can.

And be able to weight those sources appropriately
to define what a proof of definitive truth looks like about a certain set of events leading
up to a definitive truth about price or a definitive truth about a weather event. And then you also need guarantees that that
calculation will happen in a tamper-proof way. So it'll happen in a math-based way. And then you also need proof that the math-based
guarantee that you've generated from many sources, and that's computed in a math-based
provable way, is then going to be delivered correctly to the contract. Because that's where you need one math-based
system in the form of an oracle network to interface and send data to another math-based
contractually guaranteed system in the form of the smart contract. And so, basically, oracles and smart contracts
make what we call universally connected contracts. They're the more advanced version of smart
contracts. ASH BENNINGTON: Right. SERGEY NAZAROV: The solving of this problem
actually needs to happen on a case by case basis and what you really need is a highly
flexible system that can allow the inclusion of various data, can allow the various– the
application of various forms of computing and proving that the data does represent reality.

And then various ways to deliver that data,
that proof, that definitive truth, into the systems where people want to consume it. So, this means you need such a system to be
able to interact with the world's data. You need it to derive and generate definitive
truth that's cryptographically proven and guaranteed. And then you need it to be able to deliver
that into the various blockchains, Layer 2 systems. And in some case, interestingly now, we're
seeing actually web systems also want such a level of definitive truth. So, I actually think the counter-intuitive
and interesting thing is that once you do develop a level of definitive truth about
events that meets this high standard, there's actually many more use cases than smart contracts
and blockchains that want that level of assurance that an event occurred But it's smart contracts
and blockchains that have that really high standard to begin with.

So, if you build a system that meets their
extremely high standard, then offering that same level of definitive truth to other systems
with a slightly lower standard is going to be kind of easy, because you've already met
the higher standard. ASH BENNINGTON: Yeah, it sounds like you're
solving not a problem, but a stack of problems. Because you have all of these different hops,
whether it's via authentication non-repudiation at the sensor level, validating and associating
it with a particular event. Associating it then with the contract, tying
it into the network. It is a lot of different hops along the way
that need to be solved. SERGEY NAZAROV: Right, you're basically trying
to take the non-deterministic, probabilistic, unpredictable world, you're trying to feed
it through a system that creates what we call definitive truth, and what we consider to
be essentially the level of proof that a smart contract can rely on to trigger itself. And then you're feeding that into all of these
various environments.

So, you need to solve– you need to solve
a few fascinating problems. And some of them we've solved. And some of them, we're on the way to improving
the solution to. And some of them, for the more advanced versions
of an oracle network, like Chainlink, we're in the process of solving. And we recently launched a big research program.

And they're– this space, the incremental
improvements in this space, they come out of people taking genuine computer science
innovations built on the– standing on the shoulders of giants, and composing those into,
with their additional innovation, into a new infrastructure capability. And, so once that happens, you see a kind
of explosion of usage and new smart contracts being generated from that. And it's that initial– it's, I think, that
that initial spark of a new type of smart contract that we've now started lighting in
the form of DeFi, in the form of enabling a lot of DeFi in a lot of different contracts
to come into existence. And I think as we continue to research and
release new additional ways for this new oracle network form of computation to function, you'll
see more and more– for example, as we put weather data, you can have insurance.

As we put price data, you can have financial
products. As we put something called proof of reserve,
where we prove the solvency of an underlying asset as collateral, you can now use that
collateral in a financial product. And so, as we incrementally solve those problems,
you'll see more and more infrastructure appear. And you'll see more and more use cases come
out of those incremental additional variations of this infrastructure. ASH BENNINGTON: Sergey, when you talk about
definitive truth, what does that mean? And I don't mean it in an existential way,
just from a game theoretic perspective. What is the level of assurance you need? Do you think about it in terms of confidence
interval? What's the number of nines that you need to
believe that something is, in fact, true? SERGEY NAZAROV: So, I think at the end of
the day, it happens on a use-case basis, and it's kind of a spectrum of the best that you
can get. And then it's a risk assessment of how much
of this is– how much– is this a definitive truth, and how much value can I tie to that? So, we have networks right now of 21 nodes,
21 independent tamper-proof nodes, that are pulling data from 10 different data sources.

And so, those data sources prove to a certain
degree of definitive truth that that's the case. I think definitive truth has two parameters
to it, in my opinion. It has a parameter around tamper-proof-ness. It has an add a parameter around that people
can't manipulate the data to change the outcome of what the truth is. So, you actually, in my opinion, need an oracle
network. You need a system that has incorruptible,
independent computing agents to come to agreement about what definitive truth is at all. So, you need a level of tamper-proofness to
arrive at a definitive truth, in my opinion. And then from there, you start to look at
what is the dispersal of data sources? What is the dispersal of proof? And then there will be different thresholds. And it will be different in different industries. In some industries, it'll be, yeah, we had
five data sources all say to us that this is the case. Those data sources are globally distributed.

And they're all independent of each other. And I the data sources are incorruptible. I know they're independent of each other. So, at a certain point when you have enough
people telling you this is the definitive truth of the weather at this time on this
day, that is the closest you're going to get. And that's really what the system will be
able– already does, in many cases, is it gets to the closest you can to definitive
truth about certain matters. ASH BENNINGTON: Right. SERGEY NAZAROV: And those matters are– those
matters are prioritized right now against smart contract types. So, people want price data to make decentralized
financial products. So, we create definitive– framework is–
Chainlink is a framework that gets composed into oracle networks that generate that definitive
truth, for that set of contracts, about that set of financial products, such as derivatives,
or lending, or whatever. But we have others where we have oracle networks
that generate definitive truth about weather events, that on production today are actually
executing insurance contracts for emerging markets and developed markets.

And so, that's another version of definitive
truth, and their standard. And that's actually what Chainlink is meant
to do. Chainlink is meant to allow people to define–
firstly, the user can define in a service agreement what they consider to be definitive
truth and what their threshold. And then they can require that of the network. And then the network can be composed of the
node operators and data sources that go on to define definitive truth to the standard
of that user and of that contract. ASH BENNINGTON: Right. So, it's a functional and operational definition. And it's agreed upon in a service level agreement. But what makes this unique from a traditional
agreement, for example, of an insurance company, is that it's public, it's transparent, and
it's verified by the laws of math and physics.

SERGEY NAZAROV: Right, and the service level
agreement, just so you understand, it's not a paper document. You can have a paper document attached. But the service level agreement happens on
chain. So, you actually have a– you have on-chain
commitment from a system saying, I'm going to deliver data to you at a certain frequency,
at a certain quality, for a certain time period. And then– so, you have the commitment, and
you have the data also being delivered on chain. So, this means you have a lot of proof about
whether certain oracles, or certain compositions, or ensembles of oracles, or nodes, has been
successful at delivering certain types of data. Because, their commitment is memorialized. And their delivery is memorialized in this
math-based system.

And all the economics that flow to the node
operators is enforced in such a math-based system, such that if they don't deliver the
data, they could suffer a penalty, or their fees could not be released to them, or any
number of other kind of dynamics, right? ASH BENNINGTON: Right. SERGEY NAZAROV: And so, it's really this fascinating
question of how do people compose– how do people have a system that composes definitive
truth for them on a use-case by use-case basis? And also, very importantly, if the value in
the contract increases from $10 million, to $100 million, to a billion, perhaps you get
to a point where you say, hey, I want more definitive truth, I want more proof.

I'm not comfortable with just having seven
nodes asking three data sources or five data sources. I want 70 nodes asking 10 data sources. Or I want even more. Because now, what's at stake is greater. ASH BENNINGTON: Right. SERGEY NAZAROV: And I want to scale my relationship
to definitive truth about the real world in a way that relates back to the value controlled
by such a system. ASH BENNINGTON: And smart contracts can build
that scaling in programmatically. So, as the value scales up, you can have a
scaling factor on the source of truth side, as well. SERGEY NAZAROV: That's right. The smart contracts can basically say, once
I hit this threshold, I need to go request three more data sources of AAA quality. And I need to have at least– I need that
validated by at least 15 more nodes of this quality level.

And you're absolutely right. The system will get to a point where people
will be able to dynamically request more reliability, more proof. And that'll scale with the value secured. And really, if you think about it, this is
how you would– we're really talking here about making an automated world of contractual
agreements. And then the question is, how can you make
that automated world? And I think the answer is, when you can rely
on all the automated systems responsible for controlling the agreement. And one of the key parameters, or in many
cases the starting point of that, is what's going to trigger the system? And so, until you have that definitive answer
of this is the right answer to trigger the system, you can't exactly make the perfect
automated contract. ASH BENNINGTON: Right. SERGEY NAZAROV: You will always have to have
people in the mix.

Or you'll always have to have some system
that's– and that's not to say that every contract on the planet can be automated. I can't make an automated contract about I
painted your house the right color blue. Like I can't– there's always– that's another
thing people kind of misunderstand, is they think we're saying, hey, every contract ever,
forever. Did you like the TV you bought? Did it meet your expectations? Like, we can't prove that with data. ASH BENNINGTON: Right. SERGEY NAZAROV: There's a huge amount of things
that can be proven with data. And the amount of them is only going to increase. And the global financial system, and even
emerging markets, are all going to want and need a contractual system that functions in
a provable, transparent, and efficient way, almost– and really importantly actually,
regardless of what's happening in their local legal system. So, the other fascinating thing is, you now
have a parallel technologically enforced legal system that people can rely on regardless
of international dynamics or local governance structures in a certain economy.

And that's actually proven by the tokenization
movement and its global nature. Is that people all over the world can participate
and utilize this infrastructure without any consideration about the legal dynamics of
any one country or any one geography. ASH BENNINGTON: Right. And that's fascinating as well, for the reasons
that we were talking about earlier, about the international system. And people who have grown up in the United
States, we're generally accustomed to the rule of law, to the idea that contracts are
enforceable. But the idea that you could extend that through
math and physics-based proofs to the entire world, it potentially opens up a world of
possibility for international applications. SERGEY NAZAROV: Right. And international applications is one thing,
right? It doesn't necessarily– if you have a parallel
technologically enforced legal system that can guarantee to you that if you conduct some
kind of agreement with somebody on the other side of the planet, and it's all driven by
data, and it's all completely memorialized, and clearly guaranteed, and clearly proven
to work against the parameters you want, you don't need to care about their local legal
system.

You don't need to think about that, or the
relationship that they have with whatever other legal system. You can now conduct global trade, global economic
transactions of all kinds. The other thing that it does, is it actually
enables all kinds of startups, and insurance companies, and other financial firms to sidestep
nonfunctioning emerging market legal systems. So, what it allows them to do is to say, look,
if you're in a country that people have a phone, which is basically every country now,
and that phone has access to the Internet, we can offer you a savings account.

We can offer you crop insurance. We can offer you access to all of these financial
products that help you combat inflation, help you insure yourself against all kinds of risk. Because the relationship that we have with
you as a user is not based on your legal system. It's based– and the user doesn't want that
either. There are so many places in the world where
the legal system simply doesn't enable certain types of economic activity.

And I think that a parallel technologically
enforced legal system is going to just give people a better alternative. I think this technology is in the process
of getting polished, just like many other technologies that leapfrog in emerging markets
get polished. And once it gets polished to a certain threshold,
where you no longer– you no longer need to be an expert– like, once the security and
ability to build with this infrastructure is so polished and baked into it that any
developer anywhere can just spin it up and build a financial product, or insurance company,
or savings account, you will you will see the same thing that you saw with telecoms,
and cell phones, and the same thing that you saw with the Internet, and access to information,
you will see that for contractual agreements through this format. ASH BENNINGTON: Right. We often talk about scaling problems, for
example, in Bitcoin with the method of– medium of exchange function. But it strikes me that with smart contracts,
with oracles, there's potentially the reverse effect.

There's potentially a kind of Metcalfe's Law
effect, where when the number of sensors in the domain set scales linearly, the ability
to verify things will scale geometrically, so that there's a non-linear relationship. And as the system scales up, you get a greater
level of truth. SERGEY NAZAROV: Yeah, I think that's absolutely
the case. I think there's two dynamics. I think one dynamic is where we already see
many users using the same data sets that other users use.

And so, we don't need to launch the same network
again. And as we launch every additional network,
we see multiple users now showing up and using oracle networks about specific pieces of data. And the fees from those users go to support
that oracle network. So, eventually you arrive at a place where
as more users use a certain oracle network or data resource, their fees, they don't go
to us. They go to the node operators. And so, as more fees begin to accrue around
a single oracle network for a single piece of data, that is the place that has the most
fees to get the best quality data, the best quality nodes, and grow into the best resource
for data. And we're already seeing multiple users simply
sign on and start using existing oracle networks, where they contribute their user fee. That user fee is then used to buy more data
for that network, more nodes for that network. And therefore, grow the security and usefulness
of that network on the basis of usage by users.

So, the more users that use that network,
the more its security grows. Because the more user fees are then distributed
into that network's use of data, that specific one network for that one specific piece of
data. And then the second dynamic that I think is
very important is the purchasing power of data that these systems will have. I think what will happen– and you already
see this happening with insurance companies, actually. You already seeing assurance companies going
to some of their customers, and together, with a sensor partner, implementing sensors
into the infrastructure of the customer, of the insurance companies policyholder, for
them to have sensors, which then feed data to the insurance company, for the insurance
company to offer them something called parametric insurance. And the person that subsidizes this, in many
cases, the insurance company. Because they actually thrive on being able
to predict risk and predict what will happen with their policyholder. And so, I think once substantial amounts of
value begin to be triggered by a network of nodes that rely on a multitude of different
data sources, the value that those data sources provide will also scale.

And then, you'll be– the system will be in
a position to say, we have another $100 billion coming online to be secured, and the people
bringing that on, they're bringing on another million dollars in user fees. We need to now distribute that million dollars
in user fees to better secure our specific network for this specific data. Guess what– we don't have a million in user
fees to go and get people to launch more sensors, go and get people to generate more data. And I've already seen that happen in the insurance
industry. Everybody's completely happy to generate more
data as long as you pay them. And as long as you trigger more value with
data, you will have more user fees.

You will have more economics to drive towards
paying for data. And so, that eventually leads to a place where
data providers, I think, eventually, are making the majority of their revenue from triggering
systems through something like this. And I think eventually everybody is going
to want– like, if you had a choice between reliable data and unreliable data, what would
you choose? Would you just choose unreliable data, and
say, hey it'll be fine, trigger my billion dollars, and my whole company's existence
depends on this data being correct? I think that's the nuance.

And the smart contract use cases, the DeFi
users, are just the first hypersensitive group to that reliability level. Because they have to provide a certain reliability
to their users. But eventually, that level of reliability
will be superior. It will be superior to any other level of
reliability for data that triggers high value things. ASH BENNINGTON: Yeah. You shifted gears a bit there. We've talked a lot about the theoretic component
of this. And you shifted gears a little bit to the
practical and where we are.

I'm curious, what's the state of play today,
right now, with the implementation of these solutions at Chainlink and elsewhere in the
world? SERGEY NAZAROV: I think what I'm seeing now
is something I've been waiting for many years, and something we've all been working towards
in our industry for many years, and that is truly really exciting. And I feel the beginning, that the true beginning
of what people call Fourth Industrial Revolution, or what people call a reimagining of the global
financial system, or the world of contracts in general. And what I'm seeing is, I'm seeing contracts
that do something other than make tokens. So, they don't make a token in order for people
to buy and speculate on a token. They make a contractual relationship between
a user and the economic outcome of the contract. And a very simple, straightforward example
of that is lending, is the ability for users to put an asset into a lending protocol and
receive an interest rate in return, and receive interest on the asset, the collateral they
ended up putting into that lending protocol. And I'm see– after seven years of developing
smart contracts, and 10 years in the– well, 10 years- – well, a number of years– in any
case, a number of years in the blockchain industry, you basically see–
you're now seeing people make financial contracts in this format that are about monumental,
kind of foundational pieces of the global financial system.

Like, how do I earn interest on an asset? What do savings accounts do? What do bonds do? They provide a yield. Now you have contracts in this format providing
people a yield. And they're doing it amazingly well. And they're doing it– they're growing literally
from below a billion dollars to over $10 billion in less than a year of collateral and value
secured in these systems. And that rate is only increasing. And they're able to generate returns for their
users of anywhere from 2% to 8%, or actually far more in certain cases. It just depends on if you believe that rate
of return is going to persist. And, so what you're seeing for the first time
is, you're seeing people go from let's make a token and sell the token, to let's make
financial contracts about derivatives, about lending, about insurance. And let's present those financial contracts
to a user base. Which right now is a limited user base, because
you need to have something called a private key. You need to have– in order to participate
in these contracts, you need your assets in the crypto format, in a cryptocurrency or
a crypto token format, which is a universe of about $350 billion.

So, even there, I think you can see, hey,
here's $350 billion of a universe of assets in a certain format. And here's about $10 billion in this set of
financial contracts, where people are able to earn 2%, 4%, 8% interest. And once again, the uniqueness of this is
that it's not about– it's not about speculation. It's about here is a financial contract that
works better. Whereas the standard normal financial system
might give you 0%, or 0.05% interest, or if you're lucky to get it, 1% interest or something,
right? And that's probably going to persist. And inflation is probably going to happen
pretty heavily. This ecosystem can give you anywhere from
2% to 8% interest consistently. As long as you take your assets, and you turn
them into a crypto format.

So, what does that say, right? And also, you see a consistent flow of new
collateral and new value into this format, which then creates a cycle of generating even
more returns for that ecosystem. So, I think the unique thing that you're really
seeing take off this year, and this is why DeFi is exciting, is DeFi is the start– and
really, the people building this, to me, are amazing– the pioneering people, like AVA,
and Synthetics, and Yearn, and Nexis Mutual, and a lot of these other amazing users of
Chainlink. All of the ones I named are actually enabled
by our oracles to do this, by having high-quality data trigger all of these financial contracts,
such that they can exist at all, because they have highquality data triggering them. I think we're kind of at the beginning of
this. If you just do the basic arithmetic, if there's
$10 billion in DeFi now, and there's $350 billion that could be in DeFi, and you're
looking at 2% to 8% interest rates, you have two serious forces pushing things into DeFi. The first serious force is, who doesn't want
2% to 8% interest on an asset? ASH BENNINGTON: Right.

SERGEY NAZAROV: And then the second serious
force is, eventually the consumer landscape, Robinhood or whoever is going to figure out
everybody wants interest yields. If all I have to do to get somebody interest
is take their dollars, and turn it into a stable coin, and put the stable coin in a
lending protocol, and I can suddenly give them– I can suddenly give them 4%– a 4%
savings account, how do you think that's going to play out in the consumer market? And how much value– like why on Earth wouldn't
trillions of in value eventually flow into an environment– the financial markets don't
care. Like, they don't care what infrastructure
they use. They don't care if their 4% yield comes from
a database running X software, or Y software, or a blockchains, right? The important thing is two things. The important thing is that they will actually
get returns from a financial product. And the second thing that's moderately important
now, but I think is probably going to rise to a fever pitch eventually, is going to be
their ability to manage counterparty risk in relation to the financial product in which
they hold value.

ASH BENNINGTON: Yeah. SERGEY NAZAROV: And both of those properties
are going to be uniquely well represented in this DeFi smart contract, math-based guarantee
format. ASH BENNINGTON: Yeah, and especially in a–
as we're filming here in October of 2020, a 10- year treasury note yielding about 75
basis points. There's just no yield to be found in the traditional
Fiat capital market system. So, it is a compelling proposition. Sergey, I feel like you anticipated my final
question with your last answer. But as you look ahead over the next, one,
three, five years, what do you see the practical state of play evolving toward in this space? SERGEY NAZAROV: I think there's a slow scenario
and there's a fast scenario.

So, the slow scenario is that this technology
continues to get better. The infrastructure, like what we make, continues
to improve, continues to enable people to build better and better financial products. And those financial products continue to make
their way in more usable and usable forms into the hands of more and more users. Whether those users are crypto users that
have a private key and a hardware wallet, or whether existing web systems figure out
how to give their users access to crypto systems, the way that you're now seeing many people
seeking to give their traditional web users access to Bitcoin.

And so, I think there's a slow case where
the technology improves. The unique benefits of the technology, the
yield generated by the contracts, the transparency in counterparty risk all becomes highly sought
after by an astute group of early adopters that continues to grow. And that will continue on the merits of the
system. And these systems already have so much value
in them, that if good startup teams come in and build good products, they will become
successful. They will get funding. And so, this ecosystem now has reached a critical
mass, where slowly it will continue to improve and make better and better versions of itself,
and get polished to the point where it will make its way into the global financial system. Probably through interfaces like banks, and
web interfaces, and all the interfaces that also don't care where they get yield. They just care that they do.

And they just want to– they want– if their
users want to consume Bitcoin, they'll give them Bitcoin. If their users want to consume DeFi products,
they'll give them DeFi products. And so that's the slow case. And I think a lot of the DeFi protocols coming
into existence now are going to be the ones that those existing web systems end up giving
people access to. So, the protocols coming into existence now
for lending, for derivatives, for all these other things, are really forming a foundation
that I think is under-appreciated, and is going to form the foundation that– banks
and all these other people, they're not going to want to rebuild this. And they're not going to even be able to handle
it, in my opinion, in many cases. They're just going to use existing protocols,
and systems, and infrastructure, like they always do.

Because they're not software companies. They're institutions that seek to give people
financial products. ASH BENNINGTON: Right. SERGEY NAZAROV: That's the slow case, right? And the fast case is kind of a bit of a scary
world, where if the global financial system goes through a bit of a shakeup based on the
underlying solvency and the underlying counterparty risk of that system, you start to find– I
think the scary version's kind of scary because it starts to call a lot of things into question.

It will start to call into question the solvency
of institutions and things that have stood around– have been around for hundreds of
years, and have been solvent, and have been relied on– insurance plans, and banks, and
all these types of things. And all of these things get uncovered when
global financial systems simultaneously hit a really big roadblock. Like, when reality gets corrected against
underlying value, right? And that's when everybody worries about solvency,
and counterparty risk, and transparency about how these systems work. They suddenly go from zero to a fever pitch,
to like 90% of people's attention and bandwidth. You have all kinds of working groups, assessments,
hearings, decisions. And there's literally like a year where people
have political capital to make changes and to enable a better system. So, if there is a global financial system
shakeup, and there is a big sensitivity to counterparty risk that comes out of that,
then I think– I can't imagine the system people would say is better than this.

This system is purpose built. And, in fact, Bitcoin itself, many believe,
came out of the financial crisis– ASH BENNINGTON: Right. SERGEY NAZAROV: –in order to solve problems
like this. And so, I think that the purpose-built, counterparty,
risk management transparency, immutability properties of blockchains are going to become
extremely in demand, in the scenario where counterparty risk, insolvency, and a lot of
these questions become top of mind. ASH BENNINGTON: Yeah. SERGEY NAZAROV: And in that fast case– in
that fast case, you see everybody, banks, asset, managers, all kinds of financial institutions
basically realizing that if they don't provide math-based guarantees, they will lose business. Because think about it. If everybody's brand-based guarantee suddenly
diminishes to near zero value, how do people compete? ASH BENNINGTON: Yeah. SERGEY NAZAROV: If some institution over there
that somebody had money in becomes insolvent, or their counterparty risk goes sideways on
their users, and then somebody is selling you on the same idea of, like, my logo's good,
my logo's been around 100 years.

And you just heard about how something that's
100 years old is no longer solvent, it's like gibberish. It's not a pitch anymore. It's not it's not a value proposition. It's almost a negative. Like, the fact that somebody is telling you
that is almost like they don't understand what the world is about. And so, everybody, I think, in that world
is going to realize very quickly that brand-based guarantees aren't their competitive advantage
any more. Their competitive advantage is now math-based
guarantees. And guess where they're going to get their
math-based guarantees from. Guess what the fountain of math-based guarantees
is going to be. It's going to be blockchains, and smart contracts,
and oracles. And if they're going to want to even retain
their current user base– you'll just see it on quarter– like quarterly earnings calls. One quarterly earnings call, we have math-based
guarantees, we're going to win the market.

Next quarterly earnings call, every single
person, every single other bank is like, yeah, we're going to have it. By the third quarterly earnings call, if people
don't have it live, people are going to be wondering what's going to be happening with
that bank or institution. And if that's the world, then this technology,
in my opinion, gets adopted overnight. ASH BENNINGTON: Yeah. Scary and fascinating at the same time. SERGEY NAZAROV: Right, it is kind of scary. The world that that– the world we would have
to get to for that to be a scenario, for things to have to move at that speed, is indeed–
is indeed unfortunate and frightening.

But it could happen slowly. So, it can still happen slowly. I still think that– or it can happen in different
parts of the world. There could be parts of the world where–
like, you can see in certain emerging markets, people are starting to use Stablecoins. Because their local token– their local currency
isn't tokenized. Their local currency has hyperinflation they
can't combat. And, yeah, so I think we'll see probably parts
of the world where we have the fast case, hopefully not the whole world. And we'll have parts of the world where we'll
see the slow case happening. So, in either case, it's– spent a lot of
time on this, very excited by it, still very convinced. Let's put it that way. ASH BENNINGTON: Sergey, thanks for joining
us. SERGEY NAZAROV: It's my pleasure, Ash. Thank you for having me on the show. I appreciate it. Thank you. NICK CORREA: Thank you for watching this interview. This is just a taste of what we do at Real
Vision. To learn more about the complex world of finance,
business, and the global economy, click on the membership link in the description.

Give us 7 days to change your life. This will be the best dollar you'd ever invest..

You May Also Like