Sreeram Kannan - Introducing Programmable Trust + EigenLayer Roadmap

Summary

Rahul outlines his vision to build Ion Layer as a coordination engine enabling open innovation on Ethereum. He explains how Ion Layer allows permissionless programmable staking, allowing anyone to build new services like rollups, co-processors, and cryptographic methods that reuse Ethereum's security. This creates positive-sum games via innovation and coordination. Rahul argues Ion Layer's shared security model is superior and mutually beneficial. He then overviews the launch roadmap - staking is live, the full ecosystem with services but no payments/slashing will launch next, and finally payments and slashing will be added.

Takeaways

• 😊 Ianler enables permissionless programmable staking on top of Ethereum, allowing anyone to build new middleware services
• 👍 Ianler brings together stakers and operators to provide shared security across services, which is strictly better than segregated security
• 💡 Ianler serves as a coordination layer for open innovation, upon which many emergent services like rollups, co-processors, new cryptography methods etc. can be built
• 🌟 Services built on Ianler, like the data availability service Iganda, are akin to SAS services in web2, which can be combined to build end-user applications
• 😎 There are economies of scale in Ianler - applications using multiple services needn't buy separate insurance for each
• 🔒 Ianler provides elastic scaling of security via a pooled amount, allowing services to draw varying security needs over time
• ❤️ In upcoming versions, Ianler will support attributable security via pre-purchasable insurance, in addition to pooled security
• 🚀 The Ianler launch is divided into 3 stages - staking (live), full ecosystem without payments, and finally payments & slashing
• ✅ A testnet covering the full Ianler ecosystem except payments will launch soon, targeting mainnet next quarter
• 🎉 There is tremendous excitement over emergent use cases like rollups, co-processors, cryptography methods that Ianler enables

Q & A

What is the subtitle of the talk about the coordination engine for open innovation referring to?

The subtitle refers to Ian layer acting as a coordination layer that enables open innovation of new digital platforms and services, similar to how governments act as a coordination layer on top of which free market economies can thrive.

What are the two main types of positive sum games?

The two main types of positive sum games are innovation and coordination. Innovation involves creating value out of something that had little or no value before. Coordination involves multiple parties working together to create something greater than what they could create individually.

How does reaking on Ian layer allow for permissionless programmable staking?

Reaking involves staking your ETH and making additional promises or taking on additional conditions beyond just running the Ethereum protocol correctly. This allows anyone to build middleware services on top of IG layer that make use of the staked ETH, enabling permissionless innovation.

What are some examples of rollup services that can be built on IG layer?

Some examples of rollup services that can be built on IG layer include decentralized sequencing infrastructure, faster bridges between rollups and Ethereum, mempool services to handle transaction load, and watchtower services to monitor rollups for faults or fraud.

What does IG layer allow in terms of co-processors?

IG layer allows co-processors that can run computations like AI or database queries off-chain, with the inputs and outputs brought back to Ethereum. This greatly expands what can be built on Ethereum.

How does shared security provide better security than segregated security?

With shared security across services on IG layer, an attacker would need to acquire enough capital to attack the entire pooled validator set to compromise any one service. This raises the security bar compared to attacking services individually.

Outlines

00:00

😃 Introduction and Overview

The speaker introduces himself, gives an overview of the vision for the Ian layer project as the coordination engine for open innovation, and explains the concept of positive-sum games and how innovation and coordination enable value creation.

05:00

😊 Explaining Ian Layer and Reaking

The speaker explains what Ian layer and reaking is - it allows permissionless programmable staking by restaking ETH and taking on additional covenants beyond the Ethereum protocol. This allows building of new middleware services and applications.

10:02

💡 Categories of Services Using Ian Layer

The speaker categorizes the types of services that can be built on Ian layer - rollup services like data availability, sequencers, bridges etc.; co-processors to run computations and bring results on-chain; new cryptographic techniques like TEEs, proofs; and Ethereum-specific services.

15:04

🌟 Benefits of Ian Layer's Shared Security Model

The speaker explains benefits of Ian layer's shared security model - it provides stronger pooled security, attributable security via slash insurance, economies of scale, and elastic scaling to meet varying security needs.

20:05

📈 Ian Layer Launch Roadmap

The speaker outlines the launch roadmap - Stage 1 of staking is live, Stage 2 will launch the ecosystem without economic components, and Stage 3 will add payments and slashing.

Keywords

💡open innovation

The concept of allowing anyone to build new digital platforms and services that can interoperate with each other. The speaker envisions a future where modular services created through open innovation can be easily combined to create innovative applications. This forms a key part of the vision behind the Ion layer project.

💡coordination layer

A layer that brings together and coordinates different components in a system to achieve a common goal. The Ion layer aims to serve as a coordination layer that connects stakers, operators, and builders of new services using programmable staking.

💡programmable staking

The ability to stake crypto assets like ETH while simultaneously agreeing to run additional software services built on top of the base staking protocol. Ion layer adds an extra set of conditions to regular ETH staking to enable flexible "re-staking" into new services.

💡modular services

Self-contained software components that carry out specific functions and can be combined together to build applications. The speaker envisions Ion layer giving rise to many modular actively validated services (AVS) which will serve as building blocks for future crypto apps.

💡shared security

The model where the same pool of validators secures multiple different applications simultaneously through staking. This creates economies of scale compared to individual siloed security models.

💡elastic scaling

The ability to dynamically allocate more or less security to a particular application based on its needs at any given time. Ion layer's shared security model enables this by allowing services to pay for only as much insurance or slashing conditions as they require.

💡rollup services

A class of services that help scale Ethereum layer 1 by offloading computation elsewhere while still benefitting from Ethereum's security. The speaker envisions many innovative rollup services being built on Ion layer.

💡co-processors

Services that allow alternate computations to be carried out off-chain before importing verified results back to Ethereum. Examples include AI apps, private database queries, trusted hardware environments etc.

💡cryptographic proofs

Mathematically verifiable proofs that certify certain claims without revealing confidential details. The speaker suggests Ion layer could enable services providing location proofs, proofs of distinct devices used, and website certificate proofs.

💡event driven actions

Automatic actions triggered on Ethereum layer 1 in response to certain event conditions detected by Ion layer services. This helps build more robust and usable defi apps.

Highlights

Innovation is a positive-sum game where we take something and make a resource out of a non-resource

Coordination is another powerful positive-sum game where parties come together to create something greater than the sum of the parts

The government acts as a coordination layer on top of which there is a free market economy for innovation, similar to what we want to build digitally

Igan layer works with Ethereum to provide decentralized trust and allow building of arbitrary new digital platforms

Igan layer allows building of new middleware services like decentralized software as a service, catalyzing innovation

Stakers on Igan layer make additional promises and take on more conditions beyond just staking ETH to provide programmable trust

Igan layer allows building rollup services like data availability, decentralized sequencing, faster bridges to help scale Ethereum

Co-processors built on Igan layer allow running computations like AI and bringing results to Ethereum

New cryptography services like trusted execution environments, secret sharing, secure multiparty computation can be built

Location proofs, proofs of identity, proofs of distinct devices accessing a service can be built and integrated

Shared security provides better protection by requiring massive capital to attack any one service

Insurance allows both shared and attributable security - get pooled protection but also guaranteed payouts if slashed

Economies of scale - don't need separate insurance for each service, just get coverage at the Igan layer level

Elastic scaling of security - large pool of staked ETH allows services to get varying security needs met flexibly

Launching full Igan layer ecosystem soon with stakers, operators, services before adding payments and slashing

Transcripts

uh good morning everybody Welcome to The

reaking Summit it's my uh real pleasure

to welcome you all here uh I'm sham uh I

started this uh Ian layer project you

know two and a half years back and it

has been uh really exciting for us to

see all the progress from there um what

I'm going to do in today's uh talk is

try to set the context for why we doing

what we're doing what is our vision for

what we want to build and how we can all

partake in actually

okay um how we can all work together in

actually making this happen um as you

can see the subtitle of this talk is the

coordination engine for open Innovation

um really this is where uh I want to

give a little bit of uh orientation

around this uh title um to start with uh

you know if you think about it if you

zoom out there are only certain kind

kind of positive some games what's a

positivism game you know when we're

engaging in Collective action certain

kinds of games are win- win or positive

sum so that there's a net positive

created out of these games and if you

think about it fundamentally there are

only two kinds of prototypical postive

sum games number one is innovation

Innovation is when you take something

and make a resource out of a

non-resource right you can take air and

you make spectr from you take oil you

make energy you take sand and you make

silicon like these are innovations that

transform like one kind of a resource to

another you know a non-resource into a

resource really um highly positive for

everybody and Innovation is also like

this that if I have an idea and you have

an idea and we exchange it both of us

have two ideas clearly positive sum

unlike other resources which are finite

and you cannot create more of so

Innovation is one stere typical

prototypical or postive some game

there's another one which is also very

powerful coordination coordination is

when many parties come together and

create something which is greater than

the sum of the parts right if they work

together if we all work together we can

do something which is much bigger than

what we can all do just going our own

separate ways um and if you look at

these two different types of posm games

there's Echo of this structure

everywhere that you look you know in

working systems so you can think of like

you know these two structures interlay

and work together with each other in

very powerful ways in different examples

um and one example is let's say you know

you take a country like the United

States um on the you know or any you

know successful country for that matter

you'd see that basically the government

acts as like a coordination layer on top

of which there is a free market a

competitive economy which can be built

on top of it which is you know akin to

open

Innovation and what we're aspiring to

here is for to to do this for digital

platforms essentially

coordination

via bringing decentralized trust who

brings trust you know trust trust is

created through like this decentralized

Collective you know in our vision

ethereum and igen layer work together to

actually create this on top of which

anybody can build arbitrary new digital

platforms which can compose with each

other and you know in our in our

ecosystem we call this AVS you know

actively validated services or you can

also think of these as uh like a

decentralized version of software as a

service that we um we see in the cloud

okay before I go in into explaining uh

you know what all we can do with this

kind of a platform I'm going to start

with the basics like for some of you who

may not be aware uh just a couple of

minutes so I in I lay what we do is we

bring together a variety of parties so I

mentioned being a coordination layer

what does it mean to be a coordination

layer you need to bring together

different kinds of parties that work

together to actually achieve a certain

goal and in our case uh it is

mainly the first side of this is stakers

so what happens is in ethereum right

what you do is you go and stake your e

stake your e what does it mean to stake

your e you put it into a contract and

then make a promise that you will hold

to the conditions and the Covenant of

the ethereum protocol what Igan layer

does is to make

this much more expansive so we call this

reaking reaking is you stake your wreath

and then you're adding on a additional

conditions taking on additional

covenants making additional promises uh

that's what you you know it's now

popularly called reaking in fact we're

calling this the reaking summit but if

you want to be really precise you would

call it permissionless programmable

staking that's really what it is what do

I mean by that so you take the e that's

staked in ethereum and then subject

yourself to additional programmable sets

of

conditions so when you take it into I

layer you're basically saying hey I'm

going to run any kinds of new middleware

Services actively validated Services

whatever you want to call it but

essentially what what you're doing is

you're saying hey I'm I'm taking my eat

and normally when I'm staking I'm

promising that I'm running the ethereum

protocol correctly but now I'm going to

promise that I run all these Services

correctly okay um and when somebody

wants to build an ABS essentially

they're talking we're talking about

building two things

number one they can build arbitrary

software you know a container in which

they can house and deploy arbitary

software and a smart contract so I

itself is a smart contract in ethereum

but it allows anybody to build new smart

contracts that talk to the IG ler

contract any new middleware or AVS can

build a new smart contract that talks to

the a l contracts and this the AVS

contract can SP ify the payment

condition the slashing conditions and

the registration conditions who can

register how much do they get paid and

how much should they get slashed so

that's the overall structure of how you

build uh how we are able to use Aon

layer to actually take the underlying

decentralized trust from ethereum and

then Supply it to any kinds of new

middlewares or services that can then be

built on top you can think of this as

the kind of open Innovation layer

anybody can build these new new kinds of

services okay so in the last slide I

call this permissionless programmable

staking right why is it programmable

staking because you're staking and then

other people permissionless can create

these middlewares and services that can

consume your staking and then create new

kinds of services based on that so you

can think of Ian lir as being a paradigm

for programmable trust okay so you know

at the base of all of this we have the I

layer Shad security system they're

calling it you another way of thinking

about it is a Shad security system why

are we calling it Shad security the same

stake or the same pool of validators are

actually sharing that security to a

variety of different applications so

that's another like model for thinking

about this there are really two things

that power this Shad security system on

the one side we have the eat staking

people can stake eat and this provides a

certain amount of of Economic Security

Economic Security means if you know that

if your service is not run correctly you

you will be able to slash a certain

amount of

e there's also a certain amount of

decentralization you because you're

borrowing the same set of node operators

that you know run something like

ethereum you can borrow the

decentralization and this gives you a a

certain amount of collusion resistance

that these are distinct operators you

know neutral set which which is actually

participating to validate your service

so these are the two dimensions of

programmable trust that are uh created

from the I lay ecosystem and now what

can you do with this you can actually

start uh building a variety of different

things and one way to like root this

thing is to take an analogy from like

the pre crypto or the web2 world and you

know you can think of in the cloud

era the you know if if you think back to

1995 and you want to build an

application you have to build your own

like you know server stack you have to

build your own you know uh

authentication payments database

everything yourself as well as building

whatever application you want this is

what you would have done if you wanted

to do web application development in

1995 in 2023 that's not what you would

do you would go basically use a cloud

service you there is a bunch of software

as a service solutions SAS Solutions on

top like o like mongodb like um you know

stripe all these things and then you

know when you want to build an end user

application you just concatenate these

pieces correctly and then you can build

whatever application you want leading to

much higher velocity of innovation how

can we kind of see an echo of this in

the uh crypto

world so you know one can start thinking

about what kinds of you know the the

middlewares and avss the actively

validated services that can be built on

top of won layer as something Akin into

these SAS services and then end user

applications can then build on top of

these services so what I'll do next is

give you like a little bit of idea of

what kinds of services can be built on

top of wag so you can categorize them in

many different ways here are a few so

number one is rollup services like

categories of services so if you think

about the ethereum road map one of the

biggest things going on in the theum

road map is the rollup Centric road map

the idea that that is going going to be

lots of rollups these rollups offload

computation from ethereum and are able

to therefore scale the whole ethereum uh

stack and in the rollup era there's lots

of rollup adjacent services that you

know may be interesting and we're seeing

a bunch of them being built you know

we're building the first one ourselves

Igan da the data availability service

the way to think about this is when

you're offloading computation you still

need a place to publish the inputs and

outputs of said computation

you know if I publish the inputs and

outputs of the computation anybody else

can then verify that I'm doing the

computation correctly so that's called a

data availability or a data publishing

system we're building Igan da as a data

availability system on using IG layer

but there's lots of other rollup

services that we're seeing emerging in

uh on the I lay ecosystem for example uh

rollups have a single Central sequencer

which orders all the transactions can we

instead build a decentralized

sequencing infrastructure on top of Ian

layer rollups take a certain lag before

they settle onto ethereum you may want

faster Bridges and there's a variety of

different Bridges being built on Ian lir

when how to handle the me that occurs in

the um rollup ecosystem you may want to

build all kinds of interesting me

services for example I want to say build

an encrypted mol for a rollup so which

means you need a bunch of nodes these

node needs needs to participate in some

kind of threshold cryptography so that

when you send a transaction no one node

is able to actually see the transaction

it's encrypted but then after the

transaction is included then it you can

actually decrypt it so you can build me

services on Igan layer and another

category that we've seen emerge is

watchtowers you know if you have not one

or two or three optimistic rollups but

thousands of optimistic rollups which is

there we going towards you have to make

sure that there are people who are

actually watching what's going on in

these rollups and trigger a fraud alert

or a fault alert when such a thing

happens you need a neutral set of nodes

to do this so again you know a new

category that we're seeing on

ier so this is rollup Services another

category which I I'm quite excited about

personally is the family of

co-processors how do you think about a

co-processor you're sitting on ethereum

and then let's say you want to run an AI

application and then get the output of

such AI application onto ethereum this

would be an example of a co-processor

you know you on ethereum you're in the

evm programming environment but I want

to access running all kinds of other

outputs you know maybe you want to run a

Linux machine uh and a program you know

for which you made a commitment and then

you want to say that hey if you run this

program then this is the output and then

bring it all back to ium to be an

example of a co-processor you want to

run a database a SQL query on a major

database and then you want to say the

inputs the outputs of said SQL query you

want to bring it back to ethereum you

want to run like a ZK service and then

you want to bring you know the outputs

of such cryptography all of these could

be examples of co-process we're seeing

many of these uh show up on ION

layer the next category is you know new

kinds of cryptographic methods um you

know I'll talk about the ioner service

which is a new uh service that we are

building later but there are things like

trusted execution environments I want to

run like a trusted execution environment

committee a trusted execution

environment is a hardware device which

has certain kinds of you know uh there

is a little bit of trust assumption in

the manufacturer like Intel and uh AMD

and and Android all of these different

Hardware manufacturers have different te

environments but you know to be able to

access te networks on you know on

ethereum is a very interesting use case

you know things like secret sharing I

want to take a secret and encode it and

send it through the network so that

nobody has access to the secret but it's

spread all through the network um you

know more more General version of that

is the secure multiparty computation or

you know fully homomorphic encryption

we're seeing all of these new categories

emerge on igon layer um there's also

other kinds of things that one can do um

you know bring proofs of various kinds

into uh the ethereum ecosystem what

kinds of proofs am I talking about

suppose you want to know like where a

node operator is located a prove of

location you may want to get uh an

attestation that basically promises what

the prove of location of a certain uh

node operator is and and one way to do

it is have a decentralized group of

nodes which ping each other through the

native peer-to-peer Network to actually

then figure out what the Ping latencies

are you you know there are systems like

this being built proof of mashhood which

is a new kind of idea from automata

which is basically the idea that I want

to know like how many distinct devices

that you know somebody is logging in

from a distinct machine a distinct Apple

phone or a distinct Android uh you want

to have proofs of identity I want to log

into a https server and then you know

get the authenticated certificate into

ethereum you know there's a bunch of

protocols like reclaim building this um

there's also you know so all these other

services are things you would want

irrespective of the fact that these are

particularly ethereum stakers right they

need a certain amount of Economic

Security they need a certain amount of

decentralization but there's also the

fact that because we're doing restating

of e it's the ethereum block proposal

that are participating in the ecosystem

and you can start doing interesting

things on the uh on the ethereum side

for example managing me on the ethereum

L1 you can start thinking about event

driven actions whenever certain sets of

things are triggered you have to

actually you know for example whenever

there's a liquidation then that

liquidation has to be taken and these

these kinds of event driven actions for

example improve the usability of these

platforms massively because you know

imagine that like you're running a def

platform and you need to calculate the

time to uh you know you need to

calculate how much over

collateralization you you need it's

basically the time to liquidation which

is actually determining the over

collateralization factor and by reducing

the time to liquidation you can actually

get very tight systems um another system

which is you know new newly proposed is

the idea of based sequencing where like

you know from Justin Drake the idea that

ethereum L1 itself can actually do uh

ordering transactions for rollups but

when you're doing that one of the things

you may want to do is how do you get

like fast pre-confirmation and if there

is ethereum Stak by the block proposers

on ethereum then and and they're restak

on agal then you could basically start

doing things like pre-confirmation they

make a certificate that hey I am going

to include your transaction and send it

to you right away in an instant and then

later if they don't they get slashed so

these are the different examples I this

is not an exhaustive list but the but

the types of things that we starting to

see on on Ion

lay and the way we think about it is the

systems that build natively on ION layer

are like the SAS Services which means

they are infrastructured pieces and end

user applications will then concatenate

a bunch of these pieces to actually

build usable applications and we talking

about how do you take crypto to a

billion users one of the things you have

to think about is what what set of like

functionalities do they need and that's

that's where we think that IG layer will

play a role is the core functionality

layer and then applications will just

mix and match these different pieces to

then get the end us of functionality

that you want okay so that's a brief

overview of what the scope of the

project is and we're talking about to be

the coordination layer for open

Innovation this is really what we mean

many of these things we had no idea that

these could be done on I layer so these

are all emergent you know lots of people

here have actually come up with many of

these different things and it's it's

amazing for us to just sit and see that

once you allow this coordination layer

what all can then emerge out of

it okay

so um in the next couple of minutes what

I'll do is briefly touch upon what is

the fundamentals of the shad secur

system um when when people think about

reaking they're thinking about something

like hey I'm reusing the same e some

kind of Leverage or some other concept

and I just want to dispel some of these

myths here so what is the core

functionality of what is actually uh

what i l is actually doing the first

point is that Shad security is strictly

better what do I mean by that so let's

forget that we're reaking from ethereum

to ion layer let's just imagine that

inside I there's a certain amount of

each stake but it's supplied to all

these Services simultaneously right so

one way to think about it is let's say

you have $1 billion restak to 1,000

services this is One World another world

in which each service has $1 million

state which world is

better right to to attack anyone service

in the other world you just need one uh

1 million whereas to attack anyone

service B the the same pool is restak

across all these Services you need 1

billion Capital as an attacker to go and

attack any one service there is a

certain rigidity a certain hardening of

security when you pull security together

we see this all all through the place

right like this is why Nations

coordinate you know you don't have City

cities don't have armies Nations have

armies sometimes even like many nation

states coordinate to create alliances

that actually work together it's exactly

the same phenomenon Shad security is

strictly better

there is a little bit of downside in

that which is in if you had segregated

security you have something attributable

to yourself each service has that 1

million whereas in this you get a little

bit of mixing together of pooling which

is good but you know if you also wanted

attributable security what we're doing

in ier in in the upcoming version not in

the version that is already live and

launched uh but in the upcoming versions

we are working on a design where you can

also get attributable security what do I

mean by that you know if you have $1

billion stake there's potentially $1

billion to be slashed and some service

you know maybe there's a bridge which

says hey I'm very very security critical

you know if my service goes down or

something gets compromised I need at

least $100 million of insurance of the

slash uh portion so instead of taking

the slash portion right now what we do

is we just burn it like ethereum does in

our V2 what we'll do is we can actually

give you a portion of that slash funds

and the ability to buy this is called

insurance and you pre- buby it and now

you not only have the pooled security to

attack any one service you need to be

able to acquire the $1 billion of

capital but to if your service gets

attacked you you know how much you can

slash uniquely this this insurance is

not over provisioned so you it is always

guaranteed that you will be able to

slash that muchoney

so that's a superpower so you can both

get the benefits of pool security and

the benefit of attributable security you

can also start seeing that there are

economies of scale which is if you're if

you're using an application the

application is using several avss built

on ION layer then you can you don't have

to pay 5x if you're using five Services

then you don't have to buy insurance

separately for each of these five

Services you just buy insurance once on

igon lay so there is is an economy of

scale and then finally there is an

elastic scaling of security you know uh

Amazon's called ec2 elastic compute

right which is I don't know how much

compute I'm going to need I'm going to

go and buy it from a common pool and

there is randomness of how much compute

is needed by different people they go

and buy the portion of compute that they

want there is a similar phenomenon in

ier which is the elastic scaling of

security there is a large pool $1

billion or whatever amount totally

sitting as security

now whenever like a different Services

there are lots of different Services

each service needs a randomly varying

amount of security why because you know

I'm running an E2 USD bridge when

there's a E2 USD price volatility people

might want to use more of that there is

a BTC to Sol like Bridge or something

else also sitting on top of I layer then

you want different amounts of security

for each of these across time and by

having a single layer through which you

can buy sh security actually make makes

it much better okay

so um what I'm going to do from here is

just go through the

um let me just uh run this through I'm

not going to talk about all these things

um I want to basically go here talk

about our timeline of what we building

and when we're going to deploy it

um

so the um right so the earlier we had

divided the igen layer launch road map

into three different stages and stage

one was stakers stage two was going to

be operators and stage three was going

to be services and instead we've rivid

it it now you know in our current launch

plan in a different way stage one which

is already live is igen layer staking

like you can restake your e natively or

using liquid staking tokens stage two

instead of only launching ing for you

know operators what we're trying to do

is we're going to launch the entire

ecosystem IG layer you can have stakers

there's operators people can launch

Services Igan da or data availability

service all of them will go live except

the economic part payments and slashing

except the economic part everybody all

the different sites can come together

and start creating useful

Services um and in stage three we're

going to add payments and slashing so

that's our road map currently

and you know we are on the stage one is

already on the main net uh we will have

a stage two test net coming soon you

know definitely this quarter hopefully

much

earlier

um which will go on Main net next

quarter and then the stage three follows

that so that's the current launch road

map of uh Ian ler we're really excited

about all these different uh new things

that can be done across the different

sides of ecosystem stakers operators you

know um people building new avss rollups

consuming a DA lots of interesting

things happening there uh you'll hear

about some of them today um thank you so

much for listening to the first

talk