eigenraw.csv

"## Speaking style of Sreeram Kannan
## Speech pattern of Sreeram Kannan
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 coordinatEigenLayer 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 EigenLayer 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 coordinatEigenLayer
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 EigenLayer
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 InnovatEigenLayer
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 EigenLayer 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 EigenLayer 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 EigenLayer
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 coordinatEigenLayer for open
Innovation this is really what we mean
many of these things we had no idea that
these could be done on EigenLayer 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 coordinatEigenLayer
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 EigenLayer 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 EigenLayer 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 EigenLayer 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 EigenLayer 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 EigenLayer 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"
"Q: I saw Vitalik’s blog post about overloading the consensus layer, and how restaking, in his view, could pose systemic risks to Ethereum. I'm curious to hear your take on his take?
Kannan: One of the things I think he wants to kind of lay out is that, “Hey don't externalize, and don't create something that, assuming that if the protocol goes wrong, Ethereum is going to fork around it.”
I think that is a pretty reasonable position from Ethereum, that you build protocols and the protocols have to internalize social consensus rather than externalize it to Ethereum.
So I read it as to not overload Ethereum social consensus, which is used only for forking the chain. And don't assume that you can build a protocol that, and because you're too big to fail, Ethereum can fork around that. So that's how I read it.
And I think it's a pretty obvious statement in our view. But I think it has to be said, somebody has to say it, so it's good that Vitalik went out and said it.
Because what we don't want is for calls to deploy code that is not properly audited, doesn't have internal security controls, and then the Ethereum community has to now work hard to figure out how to retrieve it.
I think a lot of people after reading the article have been talking a lot about restaking risks.
I want to make it super clear: anything that restaking can do, already liquid staking can do, so I view restaking as a lesser risk than liquid staking.

Q: Can you expand on that?
Kannan: Basically, you can take a liquid staking token and then deposit it into complex DeFi protocols, or you could just deposit it into validating a new layer 2, or a new oracle or any of these things.
So anything that restaking can do, liquid staking can already do. Because you know, you have the LSD [short for liquid staking derivative] token, and you can do anything with it. And one particular thing you could do with that is, of course, go and validate another network.
So I view restaking as just one particular use case of liquid staking, but actually reducing the risk of that one particular use case.

Q: Why do you think restaking is having a moment in the news?
Kannan: I don't know. I'm glad people are talking about it. Of course, anything that adds new rewards to stakers is something interesting.
I said anything that could be done with EigenLayer could be done with LSTs, but people didn't know what to do with these LSTs.
They were doing exactly the same thing that people are doing with ether, which is lending, borrowing, the same set of DeFi parameters.
I think one thing that EigenLayer did is by creating this new category, that validation, if I can borrow the Ethereum trust network to do new things: I can build a new layer 1, I can build a new like oracle network, I can build a new data availability system, I can build any system on top of the Ethereum trust network, so it internalizes all the innovation back into Ethereum, or aggregates all the innovation back into Ethereum, rather than each innovation requiring a whole new system.
So I think that narrative is quite attractive.

Q: I was just reading the news about EIP-7514, which is a short term solution for solving the overcrowdedness of validators, by limiting entries of new validators. How does that affect an EigenLayer?
Kannan: I think mostly, it means the same thing for EigenLayer that it means for liquid staking protocols, that there is going to be a smaller rate at which new validators can enter.
There's a long entry queue right now, and people don't want to wait that long.
And making it slower is going to just make the new growth of LSTs slower. But I understand fully that this is a super important thing for Ethereum to be conservative and not have an overflow of validators that may not be able to be handled by the consensus layer.
But in the long term, if the total staking of Ethereum cannot grow, one of the things that happens is the total yield or the return that stakers are getting is bounded by the Ethereum staking, whereas in the presence of restaking there is a possibility for them to get some of these additional rewards. Other than that, it's pretty similar.

Q: You were making the point that EigenDA is just like in-house AVS (actively validated service) – explain what it is:
Kannan: What we decided is, in order to keep this system of shared security, in order to keep EigenLayer as decentralized as possible, we want to make sure that there is a highly scalable data system at its backbone. And that's what EigenDA is, it's a highly scalable data availability system, built on the same ideas that underpin the Ethereum roadmap, particularly what is called danksharding.
Our view is that building an Ethereum-adjacent data availability layer requires first principles thinking, whereas Celestia and Avail are built to be chains by themselves.
If you're building a data availability system adjacent to Ethereum, you'd want Ethereum validators to participate. So that's just one part of the story. Of course, EigenLayer enables that.
But then you go beyond that, and then you see, “Oh, it's not just you want to get the Ethereum nodes to participate.”
Ethereum already has consensus built in, and Ethereum gives you the ordering of the various transactions. So you should build the data availability system, which doesn't need its own ordering.
Whereas all the existing other protocols like Celestia and Avail, are basically chains that have to do their own ordering; we built a system which doesn't have internal ordering; all ordering is done on Ethereum.

Q: Liquid restaking tokens - once your liquid staking tokens are locked on EigenLayer, they become illiquid?
Kannan: That’s correct, the problem that the liquid restaking tokens are trying to solve is, can I just have a restaked position, and then still keep it liquid. So you can take that receipt token of liquid restaking and then transfer it.
We are not building this kind of liquid restaking but other people are building liquid restaking on top of them.

Q: I think your comment was, you want to use the Ethereum shared security for as many things as possible. I'm curious, now that there's also people building on the back of what y'all are doing, is there a natural limit to how much that you know, Ethereum can support?
Kannan: This is a similar kind of question that one could ask already at the applicatEigenLayer of Ethereum: How many applications on Ethereum are smart contracts and how many smart contracts can be built on top of Ethereum?
So it's the same thing with EigenLayer because people staking and running new applications, but now they do it much more flexibly and programmably with these aliases on top of EigenLayer, all contribute back to Ethereum. Their ETH staking increases rewards, ETH itself potentially increases in value because of all these additional use cases.
So over time, this can start to accommodate more and more.
But there's absolutely a limit."
"the most successful outcome for i l
would be web 3 offers a brand new petri
dish our job is to buy Great Tech at
great prices AI is also libertarian
enable use case that people haven't been
able to do today the next hundreds of
millions of players they will come to
the market through mobile my personal
reason why I could be bullish in the
next five months
is hey everybody welcome back to another
episode of The Block runch podcast I'm
your host Jason Choy I'm the founder of
the angel fun tangent um and everything
we discuss on this show is not Financial
advice and not reflective of our
respective company opinions now one of
the biggest busws of this year is
reaking and the go-to project in this
vertical is obviously EigenLayer now
today even before the launch of the
token Ian leer has already attracted
almost $8 billion in funds deposited
this makes it the fourth largest
protocol by total value locked according
to defi Lama so already I'm getting a
lot of messages from people saying it's
either the most most transformative
thing to happen to ethereum but also
increasingly more people are talking
about potential systematic risk that
comes with the concept of reaking so as
an angel investor in igen ler these are
the questions that I've also thought a
lot about over the past few months so
I'm really really excited and grateful
for siram the founder of EigenLayer to
come on the show to chat with us today
so welcome to the show man hey thank you
so much Jason really excited to be here
looking forward to dive into uh the
depths of the project absolutely so I
actually stayed up last night and I went
through a lot of your tweets and a lot
of the transcripts for your podcast
interviews and I realized there's
actually so so much to go over but I
guess before we dive into the meat there
i' love to kind of di the clocks back a
little bit with start with the origin
because I know you've been interested in
PP networks before 2008 uh and doing
your PhD there and then you moved on to
computational genomics I think later and
then stumbled in the crypto in 2017 and
I heard that you almost began your
journey by building on cardano first so
I'm curious you know how did you go from
that to deciding to build this kind of
reaking idea yeah no actually like we
didn't you know we didn't Begin by
trying to build on cardano uh the story
goes back even further so you know in in
2017 2018 you know around end of 2017
actually when I got to uh know about
blockchain my first reaction was you
know is this some kind of like a
speculative bubble that you know and
last time you know I worked on
peer-to-peer wireless networks that
didn't turn out well
so I I was a bit skeptical of the whole
premise because you know centralized
systems are more efficient and they have
a way to like outrun decentralized
systems so that was the kind of starting
point but you know the when it turned
for me is when we realize that uh
blockchains can basically help us
coordinate in the absence of trust and
just like the internet is the
information super iway blockchains and
crypto could be our coordination super
iway way that was the kind of like
operating model and and once I could see
that okay actually if if trust is so
Central if we can have a neutral
decentralized source of trust that can
be like really valuable but you know the
the particular problems we are working
on at that time was you know how do you
scale a consensus protocol what
properties can a consensus protocol have
you know how would you build uh data
availability or an oracle or other
systems with various properties what is
the game theory for some of these things
what are the mathematical guarantees
this was kind of how we got introduced
into it and the obvious kind of thing we
wanted to do was to get like you know
Bitcoin or ethereum to use some of these
ideas and you know we were so far away
from the space and had no real uh
connections that we found it quite
difficult to kind of understand like and
and interact with the crypto space and
uh you know it appeared that the most
kind of like uh the the only way we were
seeing that people from the academic
background where getting into blockchain
and crypto was actually going and
starting hey you know here's a new
consensus protocol I'm going to build a
layer one blockchain around it and we
thought okay there's no other way this
is the only way to like take an idea at
the level of a consensus protocol
because you can't build it as a smart
contract on ethereum because the whole
point is to change the way that the the
nodes come into
consensus and so I said okay maybe we
should build our own like you know uh
layer one and we were actually you know
we took a very purist approach and we
said like let's first actually build the
whole system before we even do a fund
raise so we had a project called
Trifecta where we built the block built
a blockchain you know we were running it
on 100 nodes at that time there's an MIT
crypto economic summit video where we
demonstrated running at like you know
hundreds of thousands of transactions
per second but that project never
launched and that's why you know if if
you're a listener and wondering where is
this trifecto blockchain it doesn't
exist it was uh
it it it it only has had a uh short
Lifeline you know lifetime as a kind of
proof of concept so I because you know
we the questions that we got asked from
investors and other people was hey okay
you you got these 100,000 transactions
per second where is the user base where
is the ecosystem how are you going to
build all of this out and I said I don't
know you know if blockchain's a thing
you know fast blockchain is a thing and
here is the thing that actually does it
and from from that point actually you
know one of the things that I realized
is kind of like a fundamental problem
for starting a new like you know decent
L protocol is the source of trust right
I couldn't just take it and throw it on
top of ethereum if it was like that that
would have been so amazing right like I
just just like you write a smart
contract and throw it on ethereum Let
the blockchain like supplies the trust
so this is when I came up with this
model that you know it's very simple
once you state it basically like a
general purpose smart contract
blockchain is like selling trust and
earning fees in return for selling trust
and selling trust to whom to
Applications right so applications are
choosing to live on top of a blockchain
because it's consuming the trust okay so
once I understood this economy the
question was like how can we do this
more generally like you know more than
just smart contracts can we kind of like
take the same set of nodes same
economics and some more transfer it and
we Tred all kinds of things can you
build it on bitcoin you know can you
build it on ethereum uh and the the
original set of ideas we had were not
really related to reaking it was some
kind of like a complex cryptoeconomic
game but the the core idea was hey now I
know I can run a different virtual
machine on top of a different different
blockchain what would I do with that and
this was you know in 2021 and one of the
things we knew we could do was to run
evm on nonm blockchains and so this was
the kind of starting point of the kind
of ion lay project is we said oh let's
bring evm to all the non evm blockchains
and you know as a starting point we we
took cardano because cardano was
probably it had a barly functional smart"
"contract programming environment at that
time and what that meant was and and it
was very difficult for developers to
build on top of it so if we had evm
easily virtualized on top of Gano that
would be like a cool thing and you could
do this on any blockchain you know it
was not specific to cardano but it was a
starting point for what we could do but
you know as we were thinking about it
one of the things that clicked to me was
if the fundamental thing is
your you want to have a common source of
decentralized trust then building on
distinct source of decentralized trust
is replicated effort rather than
aggregated effort and so the obvious
place which was powerful and Central for
us to build was ethereum and so we
wanted to then like think about what we
would build on top of ethereum and at
that time I actually had no good idea
for what to build on ethereum and the
the story is you know I had a an
interaction with uh uh Kyle samani of
multicoin and you know pitching this
this idea that you can actually do
virtualization of evm or non evm
blockchains and Kyle in his
characteristic style said you know evm
is a piece of you know star I don't want
you know you should not be working on
that and I'm like okay
and then uh you know I said oh no but
you could run like arbitary virtual
machines on top of ethereum too like if
you wanted and then he said no this is
some kind of an optimistic type rollup
it it will never work I said like why
would it work because optimistic rollups
are going to be very
expensive and I went back and we sat
down with the team and we were trying to
understand this peculiar statement I
didn't know at that time went through
all the optimistic rups this is 2021 and
found that the fees was much higher
and I thought ZK rollups should be more
expensive because I have to give a ZK
proof so why is an optimistic rollup
more expensive it was because of data
availability and all the costs were
going into publishing data on ethereum
and so we then said oh yeah you know now
it all fit together we had been doing
research on data availability for many
many years before that so we said I know
how to scale data let's just build that
so we said okay we're building on
ethereum we'll build data availability
on top of it so now we have a killer
application so that that was the
beginning of how we decided to build on
ethereum but also know we wanted to
build on etherum we just had didn't have
a good use case you know at that time
you know we didn't know what programming
environment would be interesting I had
all these like high level pictures where
I'd say oh you can build AI you can
build like databases you can build like
gaming you know environments all these
things but you know you have to start
somewhere where you know there is Market
traction and we found that with data
availability so that's the origin of Ian
da and EigenLayer on on ethereum add to
this what happened was we said like
let's go to eat Denver you know we're
going to eat Denver now next week but
this was you know
2022 and go to Denver and hang out with
all the people that I didn't know
anybody there I actually know zero
people in in ethereum at that time and I
connected with the vi of ethereum which
was you know decentralization
permissionless Innovation censorship
resistance it felt like this community
had a set of principles that they're
anchoring on rather than you know either
individuals or expediency towards profit
or even a product which felt very
different to me I came back and told a
friend who was working in Google at that
time like imagine somebody just goes to
a Google conference they're not a Google
stock owner they're not a Google like
you know programmer they're not even a
Google really like application developer
they just feel like they're a Google
person at the end of a Google conf
that's how I felt at the end ofen and so
we decided to build on ethereum after
that I yeah I think the the three
biggest lessons that I've learned over
the past years of investing in the space
is that ethereum's greatest exports are
three things right it's the liquidity of
ethereum is the programmability of if
evm and the ideology of ethereum and it
sounds like to me that when I first saw
I L I thought okay this is actually a
project that helps expand this you know
vastly beyond just the ethereum chain
itself and I know you describe I as a
generalized mechanism for anybody to
build arbitrary distributed systems on
top of the ethereum trust Network and
when my Normie Friends Ask me okay what
the hell does that mean I basically
explained to them well you can stake
your eth and not just validate for the
eth network but for you know any project
that chooses to borrow the security of
this uh you know of this stake so I was
actually quite fascinated because uh I
feel like I've seen the similar ideas
before in like polka do or Cosmos this
idea of like shared security so how much
of the inspiration came from observing
what is being done on other chains what
work what didn't work I would say that
uh at least our own Journey was uh from
we started from looking at
Bitcoin uh where we we were looking at
hey you know can you borrow trust from
Bitcoin and the set of ideas we got
exposed to was mostly merge mining the
idea that you have a mining common
mining uh Power and a lot of energy is
expended on mining can you reuse that
mining for other like securing other
blockchains this idea was called Mudge
mining back in in the day and I think
even Satoshi wrote about it in one of
their no Bitcoin talk or whatever and
the thing
is um we looked at merge Mining and one
of the big incentive problems in merge
mining is if you merge mine Bitcoin and
some other altcoin the problem is you
can attack the altcoin with impunity
because I you know if you if there is an
attack on the other chain it doesn't do
anything to my mining equipment or like
my Bitcoin price is not affected by the
attack that happens on this other chain
or coin so the crypto economics of
Bitcoin is very uniquely tied to bitcoin
itself which is that I have invested
this bunch of money in like buying the
mining hardware and therefore if I do
some big harm and a majority tries to do
an attack then my mining Hardware may
become useless you know because you know
the the the BTC is not valuable anymore
so this was not the case for you know
merge Mining and so the realization came
from when you do staking the Dynamics
are quite different because staking
comes with slashing the equivalent of
slashing is like going and finding out
which miners behave badly and then
burning their mining equipment right
like that's not even thinkable as a as
as a possible strategy right uh but the
thing with staking is stake is
intrinsically like you know on the chain
and therefore you can burn the stake for
misbehaviors on on the chain and this
was such a powerful concept and then we
realized that if you had a general
purpose staking mechanism you can
transfer the cryptoeconomic trust to
arbitary other services much more easily
we didn't know about much of the other
other stuff like poka do or even layer
tws or you know what was going on in
Cosmos or avalanch at that time but of
course later as you start exploring
these ideas you find oh yeah you know"
"these these other things are on this
vicinity and what we I think have built
is the most generalized system for shat
security which is more General than all
the other systems in in the flexibility
with which trust can be offered and
shared so that that's how I would phrase
it in relationship with some of the
other systems like any networks or
protocols there's many different
stakeholders so obviously there's rakers
who are staking their eth and then also
choosing to provide security with that
stake to other applications there's The
Operators that run the network but I
think the concept of the AVS is quite
interesting and not something that you
know many people in crypto have kind of
really come across before EigenLayer so
can you help us explain you know what
exactly are these abs maybe that's a
good SEC way to talk about the the
origins of IG da as well yeah um ABS is
actively validated services it's a term
that we co to explain what types of
things can be built on EigenLayer because a
lot of people look at EigenLayer you
know at least right now in in the
context of some kind of a D5 protocol it
was never intended to be a D5 protocol
it was intended to connect stakers and
operators to innovators people inventing
new distributed systems and
decentralized protocols to borrow and
share the the trust and security and we
had to come up with an umbrella term you
know people usually think of these as
chains but we think chain is a very
restrictive and a narrow way of thinking
uh so we think of these as services and
why service right like you know in in
the cloud there is an analogous thing
called software as a service right
software as a service is you write a
piece of software throw it on AWS or
Google cloud and let it run there and
you know as people are using the
software you know you pay for the cloud
but you also make money and if the unit
Economics work then you're actually
making money as a SAS but the most
important thing for me in the structure
of SAS and Cloud was how much open
Innovation that actually enabled because
you have the cloud and you don't have to
think about all the hardware and like
how to run it and how to scale it and
how to Prov the enough amount of like
compute to actually run your services
what happens is you have uh a very
powerful system where anybody can come
and innovate and build new SAS services
and put it on top of the internet this
led to Super specialization very very
narrow specialized SAS services that
were built and lots of lots of people
all around the world like you know in
places which could not compete on the
hardware basis could compete purely on a
software basis and this is a very
interesting important thing that
happened with the internet and so if you
think of what is the analog analogous
thing that can be done on on top of the
the crypto blockchain infrastructure
that's what an AVS an actively validated
service an actively validated service is
a service that you uh you know write and
it is anything that requires
decentralized validation right and these
decentralized validated services are
then like managed by EigenLayer to make
sure that all the node operators opt in
the enough stake is backing it like how
much particular attributable economic
commitment has been backing that service
all of this accounting is managed by Aon
layer but as a Creator as an innovator
you can just write the service and put
it on top of like uh EigenLayer so that's
the category of uh actively validated
Services it includes something like
running a new layer one chain but not
exclusively running a chain you may be
running very specific services and you
know what might that service be to take"
"you know usual examples it might be to
run an oracle which fetches data from
the internet has a group of nodes agree
that that's the correct data and then
put it on top of a blockchain it may be
a a bridge which reads data from another
chain and then moves it on top of
another chain it may be a um you know an
AI service you may be sitting on
ethereum and you want to request some AI
inputs and you know you need to run an
AI inference to actually adjust the
prices or something on top of your Unis
swap pool so these are all examples of
services there may be other like much
more nuanced and specific services that
show up on I which we already seeing for
example uh we see things like prove of
location proof of location is I want to
know where these nodes are placed can I
run a decentralized service to know the
location of these other either operators
or users in a decentralized manner and
how might the decentralized nodes know
it by sending Network latency
information like you know I send a
packet when do I receive it back if I
can receive it back within like you know
30 milliseconds it must be close to one
of my locations
and if many many nodes can do it and
parallel and triangulate where the nodes
are so that's a really interesting kind
of piece of information so prove of
location is that that's an example of a
service but we've seen all kinds of
different Services I would say like 20
categories of services that are kind of
building on top of EigenLayer so that's
the ABS actively validated service is a
category I can go into some of these
examples but that's a high level uh
overview yeah and given how diverse
these use cases can be I'm curious uh
and this is something that a lot of
people are discussing as well in terms
of the security assumptions because you
mentioned the concept of slashing so
let's say if someone is reaking for like
20 different applications one of those
applications happen to do something bad
and you know the validator has to be
slashed uh what does that process look
like and you know how do we how do we
think about the implications for the
entire kind of uh you know EigenLayer
stack the uh the way to think about it
is
slashing is encoded into smart contracts
that talk to the EigenLayer contract
whenever a Staker is opting into a new
AVS they're basically opting
specifically into an AVS contract and
the AVS contract specifies the
conditions of registration the
conditions of payment and the conditions
of slashing so basically who can come in
what's the POs to incentive what's the
negative incentive to do this stuff and
those are encoded in the AVS contracts
and so now when um when a Staker opts
into a bunch of avss what they're
actually saying is I'm opting into these
bunch of conditions of like positive and
negative incentives and I need to do
this bunch of work to actually keep up
my uh positive incentives and if the
contracts if the slashing conditions are
return in code on ethereum Smart
contracts and you know for a moment
let's assume there are no code
programming errors which we have to deal
with but you know then essentially what
you're saying is you're opting into
rigid objectively verifiable slashing
conditions which are return in code on
ethereum and what that means is if I
know that I'm running the software
correctly I won't get slashed by this
you know by this contract in fact I may
even run a piece of code we call an
anti-s slasher what an anti slasher does
is whenever I issue a signature it
checks that this signature will not
trigger slashing you know before it
issues the signature and so this kind of
an anti- slasher can be run locally and
so you know that you will not get
slashed if you actually if the contracts
are correctly return so that's the first
level which is as a Staker you're opting
into objectively verifiable rigid
conditions return as smart contracts on
ethereum so the trust model is very
similar to the other kind of trust model
when you're going and opting into a Unis
swap or a or any of these kinds of D5
protocols but to protect users even more
we have another layer of protection
because we know that code can sometimes
be buggy and we see this all the time in
crypto like sometimes you know a
protocol has a buggy code and then like
suddenly people lose their funds and
this is something we are taking a
cautious approach to and the way we take
the cous approach is by actually having
what we call a slashing veto committee
this is a committee of external parties
you know experts in kind of protocol
design who can actually vet whether the
contract triggered the slashing for a
you know on on on the actual protocol or
it was a bug that LE led to the slashing
if it is adjudicated as a bug slashing
does not happen so therefore slashing
requires two distinct things the
objective contract to trigger the
slashing and the kind of human committee
to approve it otherwise slashing is not
fulfilled so we you know in the balance
of powers between stakers and avss we
lean on protecting stakers because you
know know stakers are basically like
underwriting the system with their own
like Risk and the the guarantee we want
to give to a Staker is if you are not
malicious you will not be slashed so
slashing is there only
for absolutely attributable attributable
actions that the Staker or operator took
which are malicious not for regular
operations where they made a
configuration mistake or the the program
had a bug or anything like that so
because you know when you're building a
pon system where anybody can come in and
participate you need to protect the
system against like malicious actors INF
infiltrating the system so you need a
system of kerma like positive or
negative incentives to keep the system
going and that's what Igan does is make
sure that as a Staker you don't have any
incentive to try to like attack the
system whereas also make sure that avss
you know have no agency to attack the
system even if they put in a buggy C
firstly they to put in smart contract
code not like you know have arbitrary
adjudication conditions but smart
contract code and then even then there
is a back stop in terms of a common
slashing V committee my guess is that a
lot of people who are kind of you know
concerned about systemic risks that Ian
ler could introduce are almost confusing
the concept of reaking with the concept
of rehypothecation because they see this
concept play out a lot in defi where
somebody has a bunch of collateral and
they use that to margin to do some sort
of lending or borrowing and then they
use the stuff they borrow to margin
again and borrow more and more and more
but this is something fundamentally
different than that right so if if you
do get slash it's not like everything
just like every single app that's tied
to that St suddenly just like collaps
and and stop functioning right just to
just to make that very clear to our
listeners yeah absolutely I think there
are lots and lots of differences I think
to take the kind of comparison between
the two two things you just laid out
imagine you take uh you know people are
thinking by reaking into 100 protocols
is the same as like you know taking 100x
leverage position actually these two
concepts are not at all related and the
easiest way to see it is if you take a
100x you know margin landing position if
the market price of that asset moves 1%
you will get liquidated you will lose
your entire position whereas if I opt
into 100 protocols and I don't act
maliciously on any of them I will never
get slashed it's a completely different
thing"
"the basic principle and I think when
vitalic was discussing a recently I
think the two kind of uh areas to look
out for that he mentioned were the
security and the centralization aspect
so I think we talked a lot about the
security aspect what are some
considerations that we should have when
we think about the centralization that
Ian L might or might not introduce
yeah I think this is a much much more
nuanced topic you know I'm glad you
brought this up um there are various
layers of decentralization that you know
protocols uh you know like ethereum may
want to have and you know the most I
think direct is operator
decentralization does EigenLayer
contribute to more pressures for
operators to centralize maybe there are
only like a few operators when there are
lots and lots of services that need to
be oper opted into and you
know the answer to this is you know
in in the structure that we are building
in igon we want to minimize the pressure
to centralize so this is a kind of like
an operating principle that we are
taking in building EigenLayer and you
know if a different team was building
EigenLayer they would operate on maybe
different principles but like I said we
came into the space because of like you
know particular ularly building on
ethereum because of the shared values
and so one of the particular things we
do is can we try and encourage Services
which do not require a lot of
computational effort and this is how Ian
da the first service is built on IG on I
layer IG da is built to be horizontally
scalable which means as you increase the
number of nodes the system's performance
keeps increasing rather than the I need
to have a lot of node requirements on
each node to satisfy a certain amount of
bandwidth so for example systems like
salana scale by vertical scaling each
node needs to have more and more in
order to actually do well and IG da
scales horizontally which means the
total performance of the system is the
product of the amount of bandwidth
available in a node times the number of
nodes so you can increase it by like
increasing the performance of a node or
you can increase it by increasing the
number of nodes and because the system
is horizontally scaling decentralization
itself becomes
scalability the more nodes you have the
more bandwidth you have and therefore
you can scale so this is a principle
that we used to build
igda okay beyond that what can we do to
encourage decentralization and I think
over time what will happen is there will
be services that require more
centralized you know operations that'll
be Services which will require more
decentralized operations and you know I
give us this example this secret sharing
imagine I have a secret and I want to
store it in a decentralized network so
each node has a little bit of the secret
if all the nodes were just the same
party it was all just coinbase like
running hundreds of nodes then I don't
get any secret sharing benefit is the
same guy just like running 100 nodes and
storing portions of the secret so if I'm
running a secret sharing Network I
actually want
decentralization and so one of the
really powerful things we're building
with EigenLayer is expressivity and
flexibility for a service to specify
that they only want let's say more
decentralized operators how do they know
which operators are more decentralized
they can choose to use the Oracles of
their choosing to decide which are more
decentralized and which are more
centralized maybe something as simple as
I want to exclude all the exchange nodes
and I want to exclude all the major lstd
nodes maybe a thing that somebody wants
to do so there are lots of expressivity
in the lay platform and if decentralized
trust actually has utility which is what
we all believe you know EigenLayer creates
a market place for the decentralized
nodes to potentially even more earn more
than centralized nodes because you know
you can't go to ethereum today and say
like hey I'm going to pay a transaction
fee but this transaction fee only goes
to home stakers that's not a thing like
your transaction fee goes to whoever
picks up the transaction and like Minds
it but on IG layer you can actually do
it you can actually say like hey I only
want to build an oracle which uses the
home
stakers and you know so we will find out
the market value of decentralized trust
by actually allowing EigenLayer to exist
and our thesis is you know there is
enough interesting things to be built
that decentralized trust has a real
value and actually for the first time
very first time in ethereum
decentralized nodes could earn something
more than centralized notes till now all
is the centralized nodes are better yeah
that's actually a huge reason why I was
very excited to invest in N layer is
because one of my big thesis is I
believe crypto is the best way to create
a market for anything um and I think
this is the first market for actual
decentralized trust you can actually put
a value like a dollar amount fee value
on how much people want to pay different
type of stakers and what sort of
centralization they actually want to see
beyond the posturing you see on Twitter
because now people can put the money
with their mouth this so I'm very
excited for that um and I love to kind
of talk about the commercial aspect as
well because one Trend that I saw in the
past few years is you know projects
verticalized into their own
infrastructure so you see Taps like dydx
becoming their own chain you see some uh
you know guilds or games like Merit
Circle you know verticalized into their
own chain so it seems like the market is
almost rewarding projects for becoming
infrastructure for
verticalized to ig layer since you know
IG layer is basically telling everybody
that hey you don't need to do that you
can just simply use existing security
from ethereum so from a commercial angle
you know what drives founders of you
know abss of apps to use EigenLayer
versus becoming an L1 or L2 themselves
the way to think about it
is if you are an AVS founder what are
the choices on the table option one
build with your own like token and your
own trust Network and option to build on
top of an existing trust Network like
ethereum I and you know deploy your
service and you know in the simplest
world one would say maybe a lot of avss
would want to get started off as the
second one which is you know um use uh
ethereum and and you know find product
Market fit and then maybe like go and do
their own
thing"
"and you know maybe this analogy
that I recently came up with may be
useful imagine you go into a store you
know you go into a mall and then there
is this main store that says that hey
you have to put up a deposit you know if
you come and steal anything here you
will lose your deposit and then you know
I come in and say hey anyway the main
store you're putting up a $100 deposit
to enter why don't you make a promise
that you know with this $100 you will
not steal anything even on the other
smaller stores in the mall they say yeah
you know now it's in your control to not
steal at the mall right like it's very
different from taking a margin lending
or any other kind of like financial
position so the risk is endogeneous to
the Staker except smart contract risk
and smart contract risk is just
pervasive in all of blockchains and
that's just what it is right and even
that we are trying to build a very
cautious govern system in the beginning
over time these governance features can
be removed but that's the trade-off that
we're taking is be cautious in
protecting the stake so you know to take
another like mental model that people
have which is I think very erroneous
when 100 Protocols are sharing common
stake the model is oh you know maybe
there's no risk from the Staker side
maybe there's a risk from the protocol
side or from the AVS side and I think
this is also erroneous and the reason is
if there are 100 protocols Each of which
can sustain $1 billion staking on their
own let's say that's the 100 protocols
and so which means they're paying some
amount of fee which is sustaining that
amount of State in that platform now if
you aggregate all of this and create a
$100 billion pool this can be restak
across 100 protocols the fee is
identical to the previous world because
you know you're paying the same fee and
you're able to sustain $100 billion now
to attack any one protocol you need 100
billion doar rather than requiring $1
billion security has this nonlinearity
where the the more stake you need to do
an attack the more stake you need to
actually profit out of and Escape in
real world with all this kind of crazy
stuff becomes impossible there is no
liquidity on an exchange there is no you
know tornado or whatever to go and hial
transactions it's simply not possible to
pull off an attack Beyond a certain
scale so there's hardening of security
that actually happens at scale so that's
the the other I think model which is
missing because when people are thinking
about 100 protocols sharing the same
stake they're thinking the amount of
Stak is going to remain the same as the
number of protocols increases but that's
not the case as more and more protocol
bring more and more fees the amount of
stake will keep increasing so this is a
market equilibrium and and another
feature that we're building with IG
layer is what we call attributable
security when 100 Protocols are sharing
a common pool of let's say 100 billion
dollars of stake there may be one
protocol which says hey you know I just
not only want to have this PO olded
security but I also want to have unique
attributable security just to myself
which means even if all the protocols
get attack simultaneously I should be
able to slash and redistribute let's say
$10 billion because I'm coinbase I want
to you know be very sure I want to be
able to re you know slash and
redistribute $10 billion on my own you
can do this on EigenLayer EigenLayer
gives you an ability to express both
unique attributable security as well as
pool Security in a common system and the
power of pool security is very similar
to why like nation states have like
security functions cities don't have
security functions is because there is a
hardening of security at scale so that's
 um so one of the things we did as
you know we thought through like what
are the incentives of the F you know of
the protocols we know that crypto runs
on incentives if the incentive is not
aligned people are not going to come and
build on top of us one of the things we
did is to break the binary choice
between hey I stake my own token or I
get security from ethereum we support
natively what we call dual staking dual
staking means I as an ABS can borrow
trust from two
distinct parties one is stakers of my
own token whose whose interests are
directly aligned with like the protocols
well-being because you know they have
exposure to the token and a neutral high
value Quorum which is coming from
ethereum so you have we have this dual
Quorum model which is a very popular
model among the many AVS even if they're
launching on a single eorum to begin
with over time they have the idea to
actually build their own other Corum and
instead of forcing a binary choice where
we say Oh either you choose your your
own token Corum or you choose you know
the Eid Korum you can say oh I'm sending
80% of the fees to the E Korum and 20%
to my own Quorum today and over time I'm
going to maybe spend send more to my
Corum and less to ethereum maybe at some
point I may even send zero to the
ethereum cor and send all the value to
myself so what this means is specific
specifically if you try to use some kind
of a discounted cash flow model to try
to Value like an AVS you know uh token
you might say that the total value that
can be accured by the AVS because you
have and the thing is this decision
between how much value goes to the AVS
token versus how much value goes to eth
is decided by the avss governance which
which will be in their own native token
so the at the end of the day EigenLayer
is continuous and pure optionality like
you have the option to use e if it is
beneficial to you you have the option to
opt out if it is beneficial to you and
what this does is it makes it breaks
this binary choice and in in this world
the value of the AVS token with igon
layer is actually only greater than the
value in the absence of wagon layer
because adding an option to consume
additional Security in a way that you
can opt in and out as needed doesn't
increase your like uh you know uh cost
basis so that's the first first thing is
the dual token model basically like
completely breaks this artificial
tradeoff and makes it very very smooth
for people to like borrow as much
security as they need to keep their
platform in in continuous utility okay
so in in this case one of the downstream
questions I get is hey does it mean that
over time uh you know services will
launch on eorum and eventually just
migrate on their own and you know this
is really a question of whether igen
layer is viable not only as a
bootstrapping platform for avss but also
as a continuous service platform for
avss and so it's incumbent on us to find
ways to create synergies across these
avss in a way that they actually want to
stay rather than they are stuck with us
and you know we have this you know
entrenched Monopoly to like keep this
platform going and there are many
interesting ways we can actually do it"
"and one way you know I pointed to
examples of the cloud early on and one
of the really powerful things Amazon's
cloud is called ec2 which is elastic
Cloud compute right an elastic compute
is the idea that I can borrow as much
computer as I want and EigenLayer is
elastic scaling of security you know if
Amazon's ec2 like ion is es2 and es2 is
basically elastic scaling of security
which is you can borrow how much
security you want and so why is this
meaningful imagine a Bridge you know
who's doing a weekly transaction volume
of like you know anywhere between 10
million and 200 million okay so now how
much security do I need if I have to
provision security separately for the
bridge I need to provision worst case
200 million so I need to have 200
million of security just for my bridge
but in the EigenLayer worldview there is
this common huge pool of security I can
randomly pull the amount of security
that I want exactly like the cloud which
am across all the services creating a
big compute platform from which you can
pull the amount of compute that you want
randomly and that's exactly what happens
with I so this reduces dramatically the
cost of security because you're not over
provisioning for the worst case you're
you're consuming security just in time
how much you need this is one benefit of
EigenLayer there's also all kinds of other
benefits where what happens is if a Dap
consumes multiple EigenLayer Services I
want an AAL I want a DA I want some
other thing instead of paying for
security separately for each of these
Services they can just pay one time
because the same pole of security is
backing all of these services so there
is an economy of scale in Igan lay that
actually incentivizes services that kind
of are mutually synergistic to stay
together so these are you can create
almost what what I call like a uh X
Market bundling like instead of BU you
know somebody could have thought and
said oh let let me create a new
middleware which is an oracle plus da
plus AI together but like who knows how
to build an oracle plus da plus AI
together in today's market condition
like figuring out how to build one is
already a huge lift so the EigenLayer
allows postmarket bundling of these
Services into like useful things that
Services can consume reducing the cost
basis as well as offering a consumer
segment across these different Services
just like you go to AWS and you have a
bunch of SAS services and you just hook
into like five you know a statistic is a
typical web2 app has 15 SAS Services
integrated in the back end something
like that could happen on IG and I think
another kind of under uh explored
incentive for developers to stick around
with i l long term is the fact that it
enables kind of New Primitives to be
built and one such new primitive that I
came across on your Twitter is this idea
of like co-processors for instance um
and you kind of talked about this idea
of like intelligent defi as well which
is you know not something that at least
I don't think you can build without ion
layer so can you kind of break us down
you know what exactly is a C- processor
and what did you mean by intelligent
defi yeah um a a Cod processor is kind
of like a layer two system but Layer Two
normally you think of as like a chain so
this is one of the reasons I don't like
the chain terminology so imagine I'm
sitting on ethereum and I'm writing a
smart contract program and you know
maybe I'm on Unis Swap and and you know
one of the things I want to do is
instead of doing passive liquidity
provisioning which is I just put it into
a pool and you know fix a certain price
level at which I'm provisioning
liquidity instead I want to dynamically
move around the liquidity right but then
the question is who is doing it are you
an active participant are you a passive
participant what is going on and what
might happen is somebody may come up
with a machine learning or AI protocol
which takes the history of all the
transactions and moves around the
liquidity like provisioning Spectrum
right based on you know these hard
inputs on the blockchain imagine that I
can actually get high integrity
provision of this service which means
when the service says that this is the
right like you know if I ran this AI on
this history and this is the output I
get and it's absolutely correct if you
had this access and what you could do is
you could write you know in uh in your
D5 program that hey I'm I'm a passive
provider but I'm provisioning liquidity
to this AI protocol and you know I'm
just hands off after that point and the
AI like sits and keeps adjusting your
like you know liquidity range and this
is a really like interesting service if
it could be built but it's not possible
to write this in evm or you know you
know fit it into the small gas limit of
ethereum but if I had an EigenLayer
service that I can sit and call on on
top of like ethereum the these nodes run
the AI service offchain and you know
sign off on the in the output of this AI
service and then put it on top of
ethereum back it with a certain amount
of Economic Security now you have a
rigid input that the protocol can take
and move around this liquidity based on
that this becomes really really powerful
because now you know we called you know
this thing smart contracts right and you
know smart contracts you know as they
exist today are rigid but not that smart
not that intelligent right you know
smart means like oh is it an agent is it
like doing complex like you know
adaptations you have to write it in
simple easy code like you know in Unis
swap has this curve called XY equals K
which is like the first kind of simple
programmatic thing that one might think
of and what if instead like you had a
complex
expressivity while not giving up
rigidity or correctness how can I get
smart and you know accurate execution
simultaneously and that's what Ian layer
promises there are other Technologies
like ZK proofs which could give you this
but they're very very expensive today
you know running a ZK proof may be as
expensive as 100,000 times just running
the software you know this is this is
possible for simple applications but
when I'm when you're talking about
running complex AI this cost just like
blows up out of control whereas on IG
layer you don't incur such a cost so
that's that's the idea of cryptoeconomic
co-processors I'm going to link your
Tweet in the show notes below as well
for anyone who wants to check out this
idea because I think this is one of the
more interesting directions that dii can
take that can really Revitalize um now S
I think my last question for you just to
wrap this up is uh let's assume that
there are many parallel realities and
we're able to zoom out and see all these
realities and in one reality we have IG
layer at its most successful and then
one where everything's gone wrong
something has you horrible happen
something horrible has happened so what
would EigenLayer look like in these two
Universe like what is the most kind of
successful outcome you can see and the
least successful outcome for EigenLayer
the most successful outcome for IG layer
would be that you know it it accelerates
this whole crypto Vision which is that
we can actually enable anybody to come
and build new and interesting services
on top of a common coordinatEigenLayer"
"and so what the means is you have this
Rich world of not only like you know we
talk about end users owning assets but
also developers owning the platform or
like working on immutable platforms on
which they're building their projects so
their long-term source of sustenance is
actually like really rewarded so many
people coming and building new
interesting Services you know a rich and
vibrant economy of consumption of these
services to build things like the open
metav ver to build our decentralized AGI
to build things like you know secure uh
private homomorphic encryption all kinds
of interesting things running on top of
this common substrate so that's what the
most successful interesting world for
looks like what is the most catastrophic
thing that can happen uh we think about
this a lot you know because we want to
make sure that we maximize the
likelihood of the successful outcome and
minimize is the likelihood of the
catastrophic outcome what would a
catastrophic outcome be something got
hacked right and you know people lost
money I think that would be a
catastrophic outcome uh we we take a lot
of precaution to try to ensure that
that's very unlikely but in blockchains
you know you we don't know these are new
and new systems so there's always some
risk that you know nobody understands so
that's that's one one aspect that could
be some smart contract problems in the
uh uh in the EigenLayer ecosystem that
leads to you know some kind of loss of
funds okay the other catastrophic
outcome I can Envision is something
happening not in EigenLayer but some
like you know layers on top and one kind
of a layer on top is this
financializatEigenLayer where something
like you know there is this whole
ecosystem of liquid reaking tokens and
what they do is they take people who
have staged uh you know in the IG layer
protocol and issue like a receipt token
which is representing their position in
EigenLayer this itself I think is not at
all harmful like having a liquid
representation of your token actually
buffers the system quite heavily because
when somebody has if you don't have a
liquid reaking position what might
happen is if somebody uh has a staged
position and wants to go and
collateralize it to like do Ling or
borrowing or whatever and they get
liquidated then the only way to like
clear off their loan is to go and
withdraw from Ian lay and from ethereum
their stake and this leads to like you
know uh pressures into the protocol into
the IG protocol as well as into ethereum
itself and what a liquid reaking token
does is by issuing a fungible
representation which is a token that can
just be that change hands instead of you
know actually going and withdrawing it
actually buffers the risk out of the
deeper layers of the stack but what we
see happening which might be a risk to
the financialized infrastructure on top
is people taking leverage on top of
their liquid reaking positions you know
when people are worried about leverage
in igen lay this is the place we should
focus on is are people lending and
borrowing against their liquid reaking
tokens in a kind of unmeasured you know
way you know ideally there is no
leverage but at you know in practice
leverage should be be kept very minimal
and this is where I I would urge The
Lending protocols for example to urge
caution when estimating the value of a
liquid reaking token instead of like
pegging it at it will always be one as
to one to eat they should take cautious
approach so that people don't take
excess risk and externalize it onto
other parties who may then have lent
their eath into these Landing protocols
and so on so that's the place where I
can see some kind of financialization go
wrong this is one of the reasons we
don't necessarily either build these
liquidy staking tokens or you know
Landing platforms but we do want to urge
all the users to exercise caution at
these layers to minimize this kind of an
outcome yeah well s I'm really
appreciative of how candid you are with
this and I look forward to seeing you
make sure that the Universe we live in
is the first one and not the second one
um so thank you so much for coming on
the show and we're going to put your
socials in the show notes below as well
so for listeners at home or if you're
watching this on YouTube just click on
the drop Dr down and you can follow
instram on Twitter so thank you so much
for coming on absolutely really enjoyed
this conversation Jason hey thanks for
supporting another episode of The Block
wrench podcast if you've enjoyed this
episode please give us a festar rating
on Spotify and apple podcast or whatever
platform you're listening to this on it
really helps us a lot or if you prefer
YouTube you can subscribe to our Channel
on YouTube as well to not miss an
episode I'd love to hear from you guys
as well and I personally make sure to
read every single comment on YouTube or
tweets that are directed at me so feel
free to leave a comment there let us
know what project you want us to bring
on or what trends you want us to talk
about or tweet at me at Mr Jason Troy or
at theblock crunch on the platform
previously known as Twitter currently
known as X and thank you so much for
supporting and I'll see you in the next
episode"