EigenLayer is Making BIG Updates to Ethereum Staking! with Sreeram Kannan

to fully verifiable, fully on chain AI agents. What if there was a universal mechanism for staking and say I can now run one hundred networks on top.

started this journey in twenty thirteen. Their hardware wallets are open source and they support a variety of tokens from Bitcoin to the top all coins and even new coins such as black Rocks, Biddle and I'm personally a user of the Treasure save five. It's a beautiful device and easy to use, and Treasure offers a handholding service where they can help you on board and set up your device so you can speak directly to a Treasure representative

who can hold your hand through the process. And if you use the code thinking Crypto all one word, you can get five percent off your purchase of a Treasure device. So to learn more about Treasure and all their great devices and services, visit the link in the description. Hey, folks, welcome into the Thinking Crypto Podcast. I'm your host, Tony Edward, and joining me is Siram Kannan, who is the founder and CEO of eigen Labs. Rem great to have you.

background and then like swerved heavy into crypto. I can tell you why. But also before genomics and actually even before bitcoin, actually I was working on peer to peer wireless systems. How do you build like a wireless network where you don't have a base station, you don't have a Wi Fi access point, nothing, just do I just

stuck to each other and then create networks. So that was what I was working on, and then I moved from that to work on genomics and then at some point discovered oh my god, bitcoin ethereum, these things are happening, and so I switched back into that. So that's that's a brief history of from my background.

then we can actually create way more human coordination. So that became like my driving thesis, and as I was looking at you know, initially what we did is we said, oh, but for this technology to scale to coordination problems, not just off money or finance, but much more broadly. Right, like trust is a kind of fundamental ingredient of all of society, and so if you want to upgrade all of that, we need much more expressive, much more programmable systems.

So you need to be able to write, for example, an AI inference protocol which is on a decentralized network. You need to write like a storage network which can store patabytes of data on a decentilized network. You want to do all these things. Existing blockchains are not meant to have this level of programmability. Blockchains like Bitcoin and

etherium expanded the amount of programmability of bitcoin. Bitcoin came up with this absolute breakthrough concept of you know, money that is verifiable, only twenty one million coins, you own your coin, right, But then Ethereum took it one step further and said, hey, now on top of Ethereum, you can write programs that live and run forever in the way that you set them up without having to trust anybody. And so anybody could come in and permissionlessly come and

innovate and create new programs. You don't have to trust who they are, what their parents are, like, whether their friends with JP Morgan or not. But on their own individual basis you can say, oh, this is a good program that I actually want to invest, to trade with, to be my counterparty, to hold my funds, whatever. And so that's a really big deal that I saw. But that still didn't have enough programmability to say I want

to start a new consensus protocol. I want to change like you know, all the notes, run AI inference, or like run some other gaming virtual machine to actually run the network. It doesn't have that kind of program really, the program really is still very limited when we think of blockchains today. You know, in twenty seven when I came and people said, oh, if we can build the next Uber, next, Twitter, next, you know, Amazon on top

of this. But simply even today we are nowhere near that we've got on like major milestones as a crypto community with coin's mainstream defis going mainstream now just getting started, but the set of things you can actually build on this goes much further. And so that's why Igen layer igen in German for your own layer. Anybody should be able to come and build your own layer, Like whatever you have in your mind, you should be able to express it and build that on top of Igen layer.

So that was the kind of core idea. And so the thing we realized is as pretty much all of crypto discovered proof of steak or steaking, which is the idea that you put down your digital assets and then promise that I'm going to run like the eat or the sole network correctly. Right, Like, say you take these blockchains, promise that you're going to run them correctly, but they were still only for that one network, right, So you steake Heath and then you run Ethereum, steak Soul and

you run Salana whatever other network. Right, But what if there was a universal mechanism for staking steak heath or whatever token and say I can now run one, two, three, four, five, six, ound eight, one hundred networks on top of this. Then anybody who wants to build a new network can think of igen layer as like just like the cloud, I just take my program and upload it to the cloud

and the cloud runs everything. Right. So now instead of each person trying to build a whole new l one or a decentralized community for every single project, they have already you know, fifteen to eighteen billion dollars of steak on, I can layer. They already have the two thousand operators that exist all around the world. And then you just press like you know, take your programs, send it to

this network. All the notes will now download and run, whether it's gaming, whether it's AI, whether it's Mataurs, whatever, the new one or the new uber, whatever you want. And that gives you a level of control and program believe that simply hasn't existed in crypto before. So that's the whole idea of iron layer. Very interesting.

create new networks. Right, so you can come in and say I want a storage network on Angonler. And then when you say that, that storage network itself doesn't require all the eath notes to run your storage network. It's basically people who have registered on Igonler, they go and run it. And all that happens on Etherium is settlement, right, And what does settlement mean. Settlement is, Hey, you said you're going to run this network, and you ran it correctly,

so I'm going to pay you out. Or you said you're going to run this network and you behave maliciously either attack the network whatever thing. Therefore, I'm going to slash you, which means take away your funds. That's the only thing happening on etherium is settlement, either of payments or off like these slashing conditions. So what that means is now all these networks are not constrained by anything

related to etherium. Right, it doesn't matter what the ethereum conformation speed is, what the etherium data throughput is, computer programmability, restruction, none of that matters. All that matters is you just have the same group of stakers and operators, and that

settlement happens on ethereum. But now the set of things you can build are actually much more complex than layer one protocols because most people when they build another layer one, they just take the existing code base and say, here is just a clone of the same thing. Right in Cosmos, there are lots of al ones. They're all just the same thing, just running another instance of the same protocol.

Whereas an agon layer people build new and unique kinds of networks because that's what So this is part of our observation is people had to start a whole new layer one ecosystem for every new idea and they just fragment it into like, oh, this is file coin, file coin storage, but that's all that it does. But is it money? Is it like has the best wallet? Does it have the best like defile. It's just like five hundred things to create to solve a layer one ecosystem.

So instead we said, don't worry about all that. Build a new and productive service that grows the net market gap of crypto itself. And so that's what people do on top of Igon layer. But the layer one layer two is kind of right, but we don't use that terminology because it's confusing. This is specifically what happens.

I want my own token to secure it. Absolutely can do that. But you don't have to on day one need that in order to just launch your network. You can use you can use iagend these already there on the network. There is people staking and running operators for this. So when you come and deploy your service you can say, yes, it's running, and then you build a certain amount of market momentum before you launch a token or whatever other

things on top. So when somebody like Uber is coming in or ideally a competitor to Uber, like Uber takes right now like fifty five percent of the fees paid to like taxi drivers, just like so unfair. In the large scheme of things we in crypto space, we talk about removing the intermediary, right, and there's no intermediary like this,

like taking food. I mean the guys like sweating driving his car, and then like somebody comes and takes some favors and I mean it's just a completely different magnitude than even payments. People say Visa takes one two percent or three percent, right, and this is somebody taking fifty five percent. So when somebody else wants to come and compete and build a competitor to Uber today, what happens is you have to write like a competitor to Uber on an etherium or a Solana smart contract, and it's

just not possible. It's completely impossible to do that because Uber has you know, you have maps, you have like data storage, you need like right matching at like you know, speed of thought. Right, you can't wait, you don't want to click and wait for like, you know, twelve minutes for some finalization to happen. But that sell out of programmability that you get on Agen layer is you can

actually do like a pretty high level of programmability. In fact, what I didn't talk about earlier is we did this recent umbrella branding called eigen Cloud. And so the idea of Eigencloud is to say, hey, there is iigen layer at the base which is providing shack security. Right. It is a common security layer, and the security layer is built on top of ethereum and it inherits you know, some of the trust properties of ethereum itself, but the

layers are built on top. We built a layer called Igenda, which is a data layer, so you can write data to it at the highest throughput in all of crypto. We just launched last week on main net an upgrade which takes it to one hundred megabytes per second. It's faster than so long, it's faster than anything else out there. So that's a throughput layer. So now when you're building your applications, you don't have to worry about where do I write or publish all my data for anybody to

access it. It's already there. It's called igen Da. It's the fastest throughput in all of crypto. You can just like write data to it. And then we're building another service on top called I can Compute, where you basically say, oh, run any computation. You don't need to worry about, like how to build a decent lized network, basically abstracting more and more components inside the IGON ecosystem and then saying it take a kind of software package, give it to

I didn't compute. It will run everything. It'll put the data on iigen Da, it'll do settlement on iron layer, everything so you don't need to think through all these things, just making it easier and easier for developers to build these complex functions very interesting.

two thousand operators on igen layer. So depending on how much you're willing to pay and how much you know decentralization that you want, you can inherit that level of decentralization for your protocol. But if you if for your protocol built on top of Vining layer, you only had one operator opt in, of course, if that operator goes down then your service might go down. But one of the things that we have in addition to just decentralization

is strong economic incentives. When you write your protocol, you could say, oh, it's just one operator, but they have to put up one hundred million dollars of steak, and if they go down, they're losing one million, two million, ten million, one hundred million, and so that's going to put a lot of pressure on them to make sure they're keeping their job straight that it doesn't go down. But this level of control, we hand it to the protocols rather than we are ensuring that every operator is

going to be alive all the time or whatever. It's up to you to decide as a protocol designer. It goes back to programmability and control. Right, you can come in and say I need five hundred operators, but you have to pay five hundred x to fees of course, right, but I need one operator, but I want them to put up one hundred million dollars of steak. That's up to you. We have the operator backing. We have the

most you know, two thousand operators. We have the steak, you know, eighteen billion dollars of steak for people to operate as long as there is incentives in your protocol to do it. It could be using your own new tokens, the incentives could come from actual fees paid by people writing this new uber. All of those things actually work.

programmability is number one. Number two, Etherium is a staking network itself, right, So people stake in ethereum and then keep the network running. And so this concept of restaking people get confused. So what it is is something really simple. When you're staking on Etherium protocol, what you're saying is, hey, I'm locking up my eth and I'm going to run the Etherium protocol correctly. When you come to eigenlayer, what

you do is you say a little bit more. You say, well, I'm locking up my eath to run both the Etherium protocol and the Uber protocol and the storage product. You choose which other protocols you're opting into, so you as a staker. You're a staker, So who's the customer type is I'm already staking on Etherium? So I say, oh, instead of just taking on ethereum and promising to run the yath note software correctly, I'm going to promise to run the eath note software and a set of things.

I choose, Well, how am I opting into those things? I decide based on am I getting paid enough for the risk I'm taking? Just like anywhere else, Right, So you go and play in a DeFi protocol, you're basically saying, hey, is the risk? Is the reward worth the risk that I'm taking? But here the risk is somewhat containerized because the risk is you just need to run the note correctly. Unlike in a DeFi or like a financial protocol, you

are underwriting financial risks which are very different. Right, this is just operational risk. You just need to be not malicious, don't attack the protocol, keep your uptime promises. You're not going to lose your state. So it's a much lower category of risk than pretty much anything existing outside. But because we have the scene in which hundreds of new networks are being built, you can earn rewards from all of them and stack your yield.

And so what that does is when somebody comes and deposits eat and wants to hold eat there, they say, oh, I also want to restake on a can layer, right, So that's like a level of service that they offer.

In addition to that, we have, like there is this concept of liquid staking protocols or liquid staking tokens where what they do is this, you go and deposit the eat into that protocol, and the protocol says, We're going to take care of steaking for you, and I'm going to issue you a receipt token that you've deposited one eat into the protocol and so I meant one. We have a protocol called Renzo. We have a protocol called Ether five, which are building liquid staking on top of

igon layer. Because it's built on top of restaking, they call themselves liquid restaking just a nuance, but it's basically the same as liquid staking. And so what this does is it basically steaks and promises that like, oh, will take care of all the abstract things for you. You just deposit eat and then get this receipt token easy eat eat or whatever. So that's the ecosystem. So lots of retail people also participate in some of those protocols.

like high degree of decent lized governance. What the SEC has come out and said is people, you know, programs can create these liquids taking tokens, and so these liquids taking tokens do not come under the regulation of issuing new stock, like you know, it's not like issuing an IPO for Google or anything. So they've come and clarified that in the large scheme of things, it's an obviously correct thing that the SEC is saying it, but it was not clear, and like it opened up. When the

regulatory thing is not clear, institutions detailed. Maybe they don't care, they'll just operate, you know, all these things. But institutions are very careful and thoughtful. Institutions will say, I don't touch any lsds, they may be legally risky and all these things. With the opening up of this regulatory clarity, we expect the biggest grade of institutions, whether it's a black crock issuing an ETF, to participate in staking and

restaking because now the opportunity is open. The other big institutional aperture is the digital asset treasury companies, right, so like starting with you know Michael Saylor and the og micro strategy to now we have a whole category of etherium treasury companies. You know joel Ubin, Tom Lee, you know Kathy Woods investing in it. Like it's a big

perio thial mesting and some of these things. So it's the big trend of hey, we need to actually open up access for crypto assets to the institutional market in a way that you also get the yield of these systems because that's a big value proposition that we have in the EATH and the DeFi ecosystem where you can get yield in size. I think if you look at protocols that can actually deliver yield at like a ten twenty billion dollars of scale, there's maybe three protocols. There's

like our Waylido ey can Layer. So and we are very happy with the SEC guidance because we have the top Us protocol right in terms of the total value locked even including things like Salana right, so we have higher value locked than pretty much anything else, you know. So that basically means we have like an opportunity to engage with the broader ar you know, institutional ecosystem. Yeah.

Like that is the kind of like based portfolio. And so if you look at that and look at compared BTC versus EATH, BTZ has the upside and of course you know, getting adoption, but EAT has both the upside and has the kind of like yield, So that gives you like a base layer of like you know, return while also potentially having like the possibility of the upside.

That's one thing. The other thing which is going in favor of the EAT ecosystems is the stable coins being issued on top of etherium and all the other layer twos on top. And people think, okay, now that stable coins are issued, is the stable coin company is going to get most of the value. And there is some truth to it, because the stable coin companies are keeping the treasury another yield coming from these stable coins, so they are actually having like a pretty smooth business model today.

But as more and more, as the stable coin regulations come up, there's going to be more issuers and people are going to compete to offer the field back to the users, so that margin is going to get tinned. But also because all of these build and settle on top of ethereum, there is other really solid value propositions. For example, as stable coins go from like a single USD stable to maybe a euro stable to a you know, UoN stable and a rupee stable and like this much

broader category of assets. Because for many other countries, while today it is the USD going on chain is opening up access, right, which is a really cool thing, but for the country at the sovereign level, it's actually at risk. The risk is the rug risk from US back to them. Right. So if you're a different country and then you think about, hey, what happens if US suddenly you know, gets into a trade war or something, and then bans every USD held

by our citizens. Suddenly all your citizens liked USDC or whatever thing goes to zero. This is a huge rug risk for those countries. So the counter response that I envision is the country is going to have like their own stables on chain number one and regulate the usage of these other stables to like some kind of constraints per person holding all these things, and once you have that, you're going to see that at the center of all

these assets is a decentralized rug resistant asset EAT. So when I want to switch from a USD to a roopee or you on, you go from USD EAT EAT, you on you go USD EAT eat roopie. So this is the programmable money thesis of eth sure, which is by enabling a big programmable economy, there is a central token and currency which is programmable, decentralized, unruggable that actually gains value over time. And I think that's the thesis that we are very bullish on EAT as programmable money

and BTC as digital gold. Right, it's simple, it's gold. It's like you hold it, it's solid. We're not going to change it all that, Whereas there is a whole other element of like programmability of the system's stable coins defy other things built on top and EAT as the natural source of like decentralized asset securing or like interacting with all of that.

we want to choke the crypto markets. So I think it's just you know, so what happens is the regulatory apparatus takes time to catch up with the President's directives, right, So we're seeing, you know, in August the SEC Commissioner issuing Project Crip proved as you know, the precedent issued it like in early January, right, So there's just that lag.

And I think once that lag clears off, you're absolutely going to see this And that's really what the arbitrage already the treashery companies are playing with, right because they're saying, hey, we can engage in if you see every one of them. They're saying staking, restaking, defile every single one of them. Yeah, because that's the things that you can do and eat

at low risk compared to anywhere else. Right. So of course you can take a BTC and like do lend borrow on centralized exchanges, but the amount of like trustless financial products sitting and engage with with EAT is much larger, specifically staking, restaking at the lowest level of risk and then adding on top other defive financial positions. So I'm very confident that that's absolutely going to happen, starting with

the treasury companies and then expanding out to ETFs. And basically you will see ETFs issued different levels of like risks. Some will just say it's just a bad ATF, some will say it's a steake DTF. Somebody will s staked and the stake DTF, right, and so like you will have this like dial to tune up and down the risk for VOT curve.

chain thesis. Oh eaight is Ethereum is going to be the big chain or Solana is going to be the big chain. Big chain which congregates all the activities and apps. So that's one thesis that you see in crypto. Another thesis that you see in crypto is the app chain thesis. None of these big chains going to matter. Is basically application specific chains like BTC like hyper liquid that are going to win because they actually build the application and

they want to own the infrastructure end to end. And what i can Layer is building is for the third thesis of crypto, it's totally different from the first two thesis. It's neither the big chain nor the app chain. It is the cloud thesis of crypto, and what if the cloud thesis is actually if you go back to ninety ninety five, you will see this is the same question that people were debating about web applications. Is there going to be a hosting service where everybody hosts their website?

Is there going to be every app runs its own servers. So it's exactly these two thesies and neither of them actually correct. The correct thing was the cloud thesis, and the cloud thesis is there are layers, each layer basically doing something specialized. Like the cloud is, oh, here's a layer, we take care of all the hardware. You don't need to think about hardware. Just deploy your software here and then you can build it. But it didn't end there.

On top of the cloud. There is the biggest economy of SaaS software as a service, right, so each SaaS company. In fact, if you take the last twenty years, the most profitable investment sector, the VC investment sector is software as a service. And why is it? It's because each of them specialize in something unique. Is like Mango dB is a database company, Snowflake is a data you know, data lake, and each one you know Stripe is a payments company, and PayPal is a peer to peer payments company.

All these kinds of things just building very special specific things on top, which basically meant any application just mix and match these services so they don't have to build any of these services themselves. Okay, so that's the kind of like the what happened in the cloud, and Eigen is building for the cloud. Piesis of Crypto, which is a third and unique these I think only we are pushing forward with because we think this is what's going to happen. And so what happens is we have iigen layer,

which is basically the like the shared data center. We have the shared security layer, which is IGON layer on top of IIGON layer, just like on a cloud you have the storage services. We have Igenda, which is a data service on top of igon layers. So you have high throughput data service on top of agon layer, and then people build other services on top. We have a thriving ecosystem of one hundred plus services built on top. Okay. So now answering your question, because this context is important

to understand. Answering your question on what is the currency of payments, who does it go to? Now it depends on whether you you're building a service or using a service. So if you're building a service, then you decide what token of payment goes to that service, how that fee is routed into eat stakers and iigenstakers and your own token stakers, so you can decide all these parameters. Whereas if you're using a service, you comply with the service

economic terms. For example, in the Eigenda, which is a service we have built. Because we build the service, it is value across to the eigen tooken. The iigenstakers get the fee, some fee split goes to eat stakers, and the fee payment has to happen either in eath or igen So that's our economic terms for the eigen Da. But as you go upstream, so we have our own service I mentioned called eigen Compute coming up. It will be similar because it's a service we built and we

are going to specify the terms. But there are all kinds of other services other people have built with their own financial terms, and so part of what we're doing now is standardizing a lot of this so that it is not a headache to mix and match these services. Just like on Amazon, you can go in a single website and shop like whatever things you want. We want to have the idem cloud console where you can come and mix and match all these services.

harmed because of this something bad. So that's a feature called redistribution that went live like within the last twenty days. When you're building a service, it's not just burn the funds. You can actually if something went wrong because your your node went down, if they have staked like one hundred million, now slash one hundred million, distribute to the app. Users who got harmed or liquidated or whatever happened. So very very powerful. Primitive doesn't exist in any of crypto today.

Then we launched on Igenda, a ten x throughput increase relative to where we are at like one hundred megabytes per second. This is last week, so again ship this thing. Now looking forward to the rest of the year and next year, what is happening. We have i can compute, which is this service where you can run general purpose computation. You're not constrained by the virtual machine inside like a

block chain. People fight and debate whether EVM is better, or the Salona Sea level virtual machine is better, or some other new suiz move VM is better or whatever. And we create a VM agnostic so you can write in any program that you want, just like you would write on top of a cloud and send it to Iigencompute. This hopefully will have it on MAInet by the end of this year. So that's coming up. That's called iigen Compute. And when we're talking about iigen compute, one of the

biggest things. So if you look at cloud, the biggest demand driver is AI same thing we think of eigencloud. The biggest demand driver is going to be AI and agents. So and you cannot write an AI or an agent inside a smart contract or a blockchain today, but you can actually write a full AI agent and deploy it on eigen Compute, on Iigencloud. So what that means is for the first time blockchains, Ethereum and all other l tools and other blockchains included are going to have access

to fully fiable, fully on chain AI agents. This was a mata in the last like I think one year back, when none of the infrastructure was there and people ran fully centralized, non verifiable trust me bro agents. That's what happened in like last cycle. And then like people are like oh, then they realize, well that's what's happening, and

then everybody moved on. But when you have AI agents that are fully verifiable, just like you deposit eighteen billion dollars in the eigen layer contract, you should be able to give an AI agent tens of billions of dollars once it's trustless, secure on chain. And so that's what

we're looking forward to. So as the as we deploy i can Compute, we expect a whole category of verifiable AI and verifiable AI agents to be built on top so that's our biggest you know, that's what's coming up on the oadmap.