DeFi Risk Modelling & Management with Marijo Radman and Benedikt Eikmanns of Alterscope

Transcript

risk taking has the potential to unlock unprecented level of innovation for the whole ecosystem yet there is a tradeoff you have to do if they have a security mechanism which is wrong you're taking risk if they have a token which is forless you taking the risk if they have a governance where one guy is taking control you're taking the risk you see now the logic why it really matters where you put your security in we build a intentionally easy to use risk engine without the need of coding to enable a broader audience to build those kind of [Music] models welcome to the interop today I am

here with Mario radman and Benedict mikeman they are the co-founders of altr scope altr scope is a new product that is quantifying risk in Defi and in crypto generally they're the first risk ZK C processor and we're going to be diving into the topic of risk in crypto and specifically how reaking and New dii Primitives are introducing new forms of risks that are not being accounted for by defi users defi protocols and institutions and how ultrascope introduces a Marketplace in order to be able to properly model and quantify those risks before we get started be sure to subscribe to get notified when new episode start drop and remember none

of what we discuss here on the interrupt is investment advice and if you enjoy this content please consider staking with us you can find interupt in the active set hey guys thanks for joining me today hi Sebastian thank you for us yeah absolutely I'm I'm really interested to talk to you guys today because I think risk is a topic that is you know very s of like underappreciated in the crypto space I think people that investing in crypto are typically risk on and are looking for the high rewards but also the high risks Associated to this asset class but as more and more institutions are coming into the space now with the Bitcoin ETF that's been approved and

hopefully the ethereum ETF very soon and just broadly a uh a broader scope of products that are targeted to institutions risk management and qualifying and quantifying risk is going to be more and more important and so this is where scope uh fits very nicely in this conversation so before we get started I'd love to hear a little bit about your background and how you got interested in this topic um sure I can start so basically I came to crypto 2016 um playing around with ethereum and once defi summer popped up uh my interest went into this direction and um at one point I just decided uh anything

besides script. doesn't make any more sense and since then I'm actually full in the space a few words about myself spent most of my career in corporate development uh space working for major German corporations uh like in automotive uh energy sector and since 2018 Krypto yes I came to crypto in 2014 um my background is I'm coming more from the research side to the topic so in 2014 I started writing a paper about Bitcoin use cases interviewed the community about change tip back then they were doing micro tipping on Reddit with Bitcoin so that was super early and

that caught my attention and ever since I followed the space wrote smaller articles um for example about use cases in the area of um energy and in 2020 started my PhD at the Technical University of Munich in the area of defi platform from economics my yeah professional background I'm Management Consultant um used to work for yeah a major strategy consultancy and mainly advised around topics of emerging Technologies and mostly for governments and institutions great you guys have a a very interesting background you know coming from the corporate and and Consulting side I think for like a lot of people in crypto

you know don't don't have this sort of very uh corporate and um background at which point did you start seeing the the issue of risk management in crypto as something that was largely underserved and needed to be solved uh it was during the defi summer at that time uh we had all these protocols coming out providing yields of 30 50 60% yet no one at that time understood where these yields are coming from it was everything focused around speculation and farming and every single project was actually focused on just getting the execution done risk management was not

important at all for us it didn't make any sense because when we look at the traditional Finance space everything is actually based on risk management it's the core of the financial industry and we started to research in this area and realized actually why this is the case and U in side actually made s sense um once def5 started um everyone was focusing on execution no one was focused on risk because there was no data and those Protocols are working in a very different way than what we know from tradition Finance space and so we had one or two years of getting a lot of those protocols generating data and then end at the end of the bull cycle we

realized okay why thisis management is important it started with three capital crash ter and so on and then everyone what got aware luckily we started very early on that side and managed to get the wave and now I believe in this cycle from the begin what we see everyone is really focusing and interesting risk measurement yeah you know in crypto there there's there's been obviously like since since the beginning of crypto there have been risks and people have taken risks and and you know lots of money has been lost uh through you know whether that's like Bridge hacks or rug pulls or uh smart contract uh security issues you know I'm thinking about like

you know the billions of dollars in Bridge hacks the Dow hack um all of the rug pools we've seen in the last couple of years and it really feels like 2022 and 2023 were pivotal in terms of bringing attention to uh the risks in the space so you mentioned Luna 3 hours capital and um and and FTX why do you think these particular events or you know the companies Associated to these particular events were a catalyst in in terms of Shifting the focus and and Shifting the attention to um really properly addressing risk in the space I I believe like from the

beginning everything was focused on security there were many auditing companies and everyone was focused getting those protocol safe and working properly risk management in general around liquidity liquidity was not that relevant and the reason was there was not much liquidity into the space put and um once the liquidity came once T went over 50 billion in tvl people started to care about liquidity related risk management besides security and then due to the composability of the space the attention went more and more

in this direction I would add there also with the Advent of Institutions coming to the space and in terms of Institutions I mean anybody who's taking from a third party money to invest it uh so when you're talking about hedge funds for example they came quite early in the space in 2021 started and for them it's of course more relevant to show that they Implement a proper risk management when you play with your own money it's not so relevant and um when you just speculating but with these kind of new players risk management becomes more and more relevant uh especially the more it's going to the

more traditional players right okay so you from what I'm hearing here it's a sort of a shift in focus from only the security and the platform risk to uh a uh being able to quantify risk uh of the associated to liquidity itself so like the quality of collateral for example or uh the risks associate to certain types of like say stable coin or um or or asset back loans this is specifically the shift that we're entering now in terms of quantifying risk in defi that's is that accurate I would say it I would refas it a little bit differently like I would say like

due to the liquidity which came to the space all areas around security and risk got magnified that means once the liquidity is there now economics matters governance matters the community matters um the compossibility interconnectedness matters because so much is at stake and when you put this B set your capital or your money into the space then you care about everything and that's basically how we also Define risk it's a more broader approach to assessing with what are you dealing there not only around liquidity and not

only around security so how do we price risk in web 3 what are the sort of best models that we have uh to accurately um assess the the risk associated to liquidity but then also uh price that risk you know when it comes to you know the yields that one gets you know at the end of the day right it's like a user uh providing liquidity in a protocol or lending against their assets and there's going to be a price Associated to that um H how does that how does that trickle down to the user there are few different elements which has to be taken in account the first one is basically you have to ensure that you are comparing apples

with apples uh for example just starting with the yield you mentioned it is important to understand how those yields are calculated where are they coming from this is the first thing then the second thing is basically security or the fundamentals of the project it's also important that to compare apples with apples so Landing protocols to Landing protocols dexes to dexes ethereum basic ecosystem should have a different assessment than for example a cosmos based ecosystem and lastly once you do these divide divisions then you have to see what the focus is like if we're talking about Landing protocols LTV um play some major

role uh if we are talking about Texas volume trading volume is extremely important for the risk calculation the uh trading volume stability is also extremely important and those elements play a crucial role and you see now already here the the issue all these Protocols are introduced in different methodologies and the understanding of mine and Benedict is also different because we have a different appetite of risk uh in space Maybe to add there I think the key theme in crypto is there are no standards yet how to assess risk when you're talking about traditional Finance

they are well established standards and practices how to analyze RIS risk how to price risk in crypto we have a completely new set a completely completely new space and as marus explained there are so many different kind of components and we often refer to this kind of NE risk space as a fundamental risk which are associated with protocols and assets and that's completely unclear how to analyze and how to assess those risks so our approach is hey we give everybody the tools that everybody can start establishing these standards start playing around with a different kind of

models and back test those so it's a completely new approach how to analyze risk and how to deal with risk and that that's what we're trying to do we're trying to establish this new base layer for risk management in crypto yeah in traditional Finance we don't have this protocol risk at least the the underlying technology risk is I think largely not taken into account because at the end of the day like there will always be sort of a legal system that will be able to come in and you know litigate on on who gets what assets if let's say things things were hacked or like rolling back transactions um with like partner Banks and clearing houses and things like that

I I I mean correct me if I'm wrong but the protocol risk aspect or at least the technology risk aspect traditional finances a lot less uh weighted than in crypto and it's largely an issue of quality of liquidity correct I would say like in traditional Finance space the best analogy would be counterparty risk you are having and protocol risk is of course something different it's focused on the technology but the base idea is quite close to to the counterparty risks you have to understand the risk interacting with this protocol before you make your decision on the

side yeah I mean it's a shift from one counterparty to the tech stake kind of so we refer to that as um also kyp know your protocol because from the counterparty it's shifting to the protocol and Tech St itself and exactly as you said Sebastian in the traditional world we have like an established legal system you have security law for example and there it's made sure that it's hard to yeah manipulate something and in crypto you don't have that and that's why we need a new approach to analyze these fundamental risk so I'd like to talk a little bit about the the product and

understanding the different components of ultrascope and also the architecture so you describe yourself as a risk ZK co-processor uh that's a mouthful can you uh break down what that means exactly and how ALR scope interacts with protocols yes so basically once we started to build our architecture and infrastructure we realize basic uh realize that it can only be used by um institutions asset managers cryp uh crypto dents to call it like that people who provide liquidity yet not really by the protocols in

itself um because uh when you're building a for example cloud-based infrastructure which is now the state-ofthe-art building centralized platforms you are taking a lot of control uh around this platform so basically we are controlling the data we are controlling the computation uh and we are controlling where the data and how the data should be provided to someone and for protocols and the at of this is not acceptable uh because protocols striving to be decentralized um having governance structures in place and they they are aware of that that there are incentives for us to exploit the protocol itself even we want to provide risk management

so for example if we are having a for example simple case customer like a who has tvl higher than 10 billion we have a quite High incentive to EXP Dey the protocol if we are able to do that and that's the reason why we need verification um and control management once we started to realize how to decentralize and optimize the Tex de we realized consensus based systems are not the way forward because they are not scaling due to the amount of data needed for risk management and with zero knowledge proof technology this problem is solved not only from the scalability but also privacy side so basically we can in a

very elegent way enable verified data sets and verified computations to enable those protocols to use uh our models on our platform and that's the whole idea of the C processor enabling protocols that they are getting smarter that they are datadriven intelligent to to be able to adjust to sudden changes in the space yet having the guarantees that we are not able to manipulate their underlying logic what are the what are the data sets that go into building a model so you know consider you know your your a

uh lending protocol for instance uh if one wanted to build a risk model around that particular D5 protocol what data points would go into the model and I guess like there's a broader question here which is also the you know I think there there's like known knowns but there might also be Known Unknown um uh data points that go into calculating risk and so how do those also get um taken into account in the model this is now the a very interesting question the qu the first question should be like what do you want to optimize on a in itself are these the economic parameters do you want to

understand the exposures you have to various other protocols in the space do you want to understand how the oracles can impact you um and when you're taking all this elements together you realize of course onchain data is is space for everything and we are running the nodes on ethereum and other major ecosystems like arbitrum optimism and so on by ourself to be sure that data quality is ensured but a lot of offchain data matters in this P because offchain data enriches the onchain data you can build more clearer and uh more advanced modeling and onch offchain data of

relevance is of course security data for example is offchain data but also like as we are in crypto and we have these communities even if something happens it can be very fast magnified Amplified through those communities so basically data from those communities which is also often plays a major role from from the Twitter Community Reddit Community Discord community and so on now the question is like what do you do want to a if you want to describe a Oracle then you want to take care about the data feits you want to take take care about the protocol itself but you want to understand also the governance structure which is in place there because if there

is a centralized Oracle indirectly they are having a control over your protocol uh if they are implemented um if we are talking about economic parameters these are mostly like onine related uh data points um and you want to understand the exposure of those uh liquidity or or over the EOS system and then possible implications or triggering Elements which could come from for example communities like when there is a speculation wave coming and so on um so first question would be like what do you want to aave on the landing protocol and then you go in this direction like our belief is like dealing with risk on for

example Landing protocols one model enough one model alone is not solving the issue you have to have many different models from the fundamental protocol risk to understand the fundamentals of the protocols you are interacting with up to the liquidity related models up to the governance related models and put all this stuff together to have a broader picture what is going on there so I'd like to zoom into this this oral uh risk for a moment because I think that's one of the most important aspects of defi is like the Reliance on on external oracles um and there's at least in the cosmos EOS system there is this uh this

notion that app chains should run their own oracles if they want to have the most reliable data and uh and also the quickest sort of data access as possible um so if we if we sort of uh look at this from the perspective of ethereum and the Reliance on protocols like like chain analysis uh what are the specific kinds of data points that go into qu like onchain and offchain that go into qualifying like the risk of using a particular Oracle I mean you have you always start when you are assessing a protocol in general with security you want to see this project is proper audited proper

checked uh you run static assessments of the code quality uh to see if best practices are implemented for example taking SBC registers as a is a reference and once you get a picture around security you go in the direction to understand the government even if the protocol is built properly you want to be sure that there is some kind of governance structure in place and the protocol does not rely on on the founders in itself um or if this is the case then it's just a centralized uh prod just just just on that point so does does

would would one have to like because I mean there's different layers here you can say okay there's like a governance protocol uh that enables the protocol to be governed by the community um but then if you take a a step further there's like okay who are the token holders that are voting on that are is the token distribution um is it is it a high token distribution or is it concentrated within a certain amount of of people and then I think at some point you also get into okay what are the what is the quality uh or what is the quality of token holders in terms of like how much they vote and here you could sort of get that data on chain but then the model I think I think and this is where I think like it starts getting a little fuzzy in

terms of the quality of the data is okay are the people who hold the tokens do they have competence to be able to make accurate decisions or good decisions for the protocol and this is kind of like this this um qualitative data that I'm not exactly sure how that gets integrated into the model is that like would that then constitute some sort of like on offchain analysis um conducted by like a third party that that would evaluate the the quality of the uh of the outcomes for the votes like this is just one example but I think there's a lot of examples here where you start getting into this qualitative data um problem that becomes difficult to

quantify um in like a ZK co-processor yes I mean zik coprocessor helps you to verify the data sets that they are immutable but the issue what you are explaining is basically what is the quality of the governance you have in place and what we are enabling is basically to our Frameworks and that was the idea from the beginning that you can optimize adjust and build the models for your own needs this is a very good example what you uh gave Sebastian you can for example take the governance structure of for example chain link and realize basically there are three wallets they they have most of the government's power in the voting uh amount of tokens when you look how the

voting uh is buil up and when you look how often they are voting so basically you realize okay most of the votes or decision making is done by them what you can then do you can build risk models about those wallets trying to understand how they are behaving what is going on you are now limited hereby because you have only onchain data you see their transactions but this is the beauty of our platform you have also a lot of offchain context to all data so basically you can see you see the wallet how he's voting how his be behaving how he's doing his trade and in parallel you can check what is going for example in the community on the social media if there is some kind of heck happening and then you just check how

this wallet behaves to those kind of events and then your bu building actually Behavior models around those user you're getting a score you're getting slowly there um at one point like you are I mean you are always limited on that side because you can just do as much as much you data have but what we hope those protocol will have an interesting interest to make their governance structure as uh transparent as possible that they will say like these major token holders should basically uh give information on them themselves who are they that people can get more deta detailed or related models around that and we then get the

quality of the governance uh in place we believe this is this is the next step of the evolution of everything but now we just have to get there that we have to um models around wallets models around protocols model on liquidity and then all this stuff will slowly get into this direction so so the so I I guess my my question here is then yeah let's maybe take a step back on the product this who are the different uh participants in the risk Marketplace uh who are providing the data and who's consuming the data I mean uh we build the whole platform in a way that you have data

providers like us who provide onchain and offchain data but we enable also thirdparty data providers to provide data this data whatever they are putting there are going to be automatically verified so basically that in case if they try to attack the protocol we can see what kind of data they provided what happens and they have to stake so basically that in the case um if they attack the vector they can lose something the interesting part is more of who is going to build those models and of course we have our own research team building those models but we build our intentionally easy to use

risk engine without the need of coding to enable a broader audience to build those kind of models so basically uh inviting researchers analysts protocal developers universities students um researchers and so on to build those kind of models who have domain related expertise in oura um to make this model much as possible a good example for example would be if you want to build a qualitative security related model there is not better place then to look for auditing companies like we can bring our own knowledge but they are doing this on a daily base it's not really disrupting their business model

but it helps for the whole ecosystem in general if we have those kind of models and then you have on the other side of consumers who are basically using those kind of models consuming the updates coming from those models on a daily hourly minute level defined uh way um and out of that you get a fly build so basically on one side you have the providers you have the consumers and on top them you have the creators um and for everything to work the creators have to have a incentive mechanism to do that so basically there is a revenue share for the model which is built and for every transaction executed they are

getting part of the transaction going into their wallet so basically the incentive is very high to build something which is useful for someone to utilize that um and that's the whole dynamics of our platform okay so you have uh data data contributors or providers so these would be like you know the Kos glass nodes token terminal they're providing uh onchain and offchain data to the marketplace then you have consumers that would be um protocols but also institutions hedge funds and and applications and I want to talk about those in a minute but then on on top of that you have this kind of

community of of people that are utilizing the data to Pro provide and create the models they're incentivized to create good models that are going to be used by protocols is is the is the incentive here uh do protocols and folks that use the models basically they're paying they're incentivizing directly the model creators uh say the community for having provided good data models or are there other sort of like rating systems uh Beyond just simply uh an economic incentive that go into um the marketplace you have always the reputation of course once you build models people start start following so

basic you get influential but the base idea is basically that you have an economic incentive to create something of value for someone to utilize that um besides the data providers I just want to add that we have also two other elements which are extremely important to mention that we have also data agents enabled so basically you can run data agents to get offchain data through our nodes additionally to complement these data companies coming on our platform and besides that service providers for example good example are monitoring companies like the companies like of cyers hyper native cube3 and so on they

are providing realtime monitoring of security for example of those protocols these are extremely interesting uh triggering Elements which can be implemented into the models uh and due that they should also be part of of the whole platform to extend the capabilities of those models it should not only be based on poor data modeling because when you're doing mod modeling you don't see a sudden change sudden shift someone H some some exploit something and those companies seeing that and this information is extremely uh valuable and if you if I would build my model I would always take those kind of information into account now on the on on the consumer side I think what's

interesting here is that so you can have a a like a passive consumption of the data data and like an active consumption as far as I understand so like for example you would have an institution right like that institution might consume the data analyze the data and then make some decision or make some investment based on um on the the data provided by the model but because we're talking about defi here and you can sort of compose things the the data can also be integrated directly into defi protocols and the yields and risks uh assoc iated to specific letting protocols or like liquidity pools or um basically defi protocols can be

automatically adjusted for the risk is that is that possible and how um how would like then ultrascope kind of uh risk data go into actively uh managing or actively informing the yield in different def5 protocols that's where the co- processor coming so so basically we are having on our platform a option that you can verify the computation and verified data sets are anyway verified automatically and with that you are getting now outputs which you can in a standard way as oracles basically

provideed data implementing your protocol but you are ensured that everything is basically verified and immutable uh on the institutional side this is less of every maybe B can give more information like how they're utilizing it's a very different way of utilization and through that the cost of data is also a different one than when looking institutions and protocols yeah when it comes to institutions this kind of automatization doesn't happen at least not yet usually they're sitting an analyst in between doing some analysis um for example setting up some kind of risk um metrics um they usually we have some committees where like an analyst provides their

work and then they decide on for example an investment strategy in this committee and then they implement the strategy so it's much more manual um also much more people are involved and it's much more yeah people driven I would say um but it obviously always depends on the type of institution we're talking about um and also Alo the use case of course because there's a broad set of use cases also when it comes to risk management can can you elaborate on those use cases so one use case is for example um admission standards so um

institutions custodians Etc they need to make decisions which assets which protocols they can interact with or which assets they can buy and which one once they refrain from using or utilizing so one use case is simple one to set up with our solution risk scores which reflect the um yeah sensitivity to risk exposure they would like to have um for example with our solution you can set set up easy um rating systems from an A C D of a protocol or like a composition of protocols and thereby

make decisions for example you go only to B everything below you can't touch as an institution so that's that's one simple use case but we going Beyond um also where we're heading to is that we enable portfolio systems so usually you don't invest just in one asset but in a portfolio of different kind of assets and we mentioned earlier for example the use case of um the case of oracles and for an institution like one core rule is in in finance to diversify so now we have this new set of risk this technical fundamental risk and when you think for example everything is

secured by one Oracle in your portfolio then you might have a systemic risk in your portfolio and that's also what we're enabling we're providing transparency we're also implementing um this use case especially in our PL PL form um to also go for example to portfolio managers and enable them to make sophisticated decisions and how to compose their portfolio on the technical fundamental level yeah I can see how this would be really useful in defi as well for like the the sort of um digital asset manager uh platform like something like somier

or quazar where you know essentially you're you're uh um delegating the portfolio cre management and the asset management to an external party and having a a a risk profile Associated to each type of asset or each pool you might bundle those into a basket that is generating yield and being able to attribute some kind of risk rating system it could be you know it could be sort of um simplified in a sort of similar rating system as we have in trafi like AAA ablea Etc and having that being integrated directly into the protocol and and perhaps also into um the yields that you get so like if

you're for example in a really high-risk Vault you might get some additional incentives for that I think that there's lots of really interesting use cases here are those things that you're also thinking about with partners and def5 protocols that you're working with yes of course that's the whole base of what we're doing you have uh walls and they basically are the Risk Managers um and they need a source to build their own risk related strategies and we are the source and not only that they are having all the tools available they have also a community which build models and they can take that and build on top of that more advanced models where they are ensured that the data

computation everything is on a proper quality level um on that side yeah I think it introduces more granularity and transparency also in terms of risks um yeah I I don't really know what goes into determining uh incentive yields for you know different pools and protocols um I I guess there must be some kind of modeling that goes into that but my my guess is that it's mostly you know driven by governance and somewhat um arbitrary decisions and introducing something like this makes it a little bit more systemic I guess uh yeah I mean you

have when justic yield as a simple example you have always the base yield the reward yield and the farming incentives campaigns coming together and you want to understand these kind of elements because if there is a farming incentive uh ongoing it is for a short time period it matters for you um and the second part of course is also you have to take this in account that uh not everyone is calculating the yield overall yield in the same way there is a mismatch there is no standardness B mentioned before uh how to do that and even when you see on the front ends those yields you have to double check the calculation which is no one doing

that this is really the yield you are uh getting um out of it I'd like to move on to to reaking because I think this is one of the really important parts of the protocol and and also the shift towards a new paradigm for defi how does reaking uh what what new risks does reaking introduce into Defi and how does ultrascope address those I mean risk taking has the potential to basically unlock unprecented level of innovation for the whole ecosystem yet there is a tradeoff you have to do

you're introducing more systemic risk in the beginning due to the node operators of ethereum for example they have to restake their ethereums on I to participate there to validate the provide Economic Security and now the whole the whole story starts there who are actually the main uh players there we have on one side the the LTS who are getting this um ethereum from those node node operators then you have the node operators in itself and you have in a very simple terms like a actively validated Services um and of course in

between is I layer when you're now looking at that what are the risks and what's happening at the moment in the space on one side you have those liquid uh risk taking protocols they are taking is etherum they're using this ethereum as collateral creating a stable coin and bringing then this stable coin into the def ecosystem so basically a lot of liquidity a lot of Engagement is happening in defy space basically they are um making defy again live um yet there is uh another element like those actively reled services are needing security for their rollup for their co-processor for theability layer

and they're using those validators for that the issue here is um at one point those uh actively validated services will enable slashing and this slashing mechanisms may lead that if the validator doesn't behave in a way they defined in a logic by themselves uh this eum can be in worst case lost this provided and this has done consequences on the LT so basically the collateral of the the peg of the stable coin is automatically affected so basically it gets automatically much more validated volatile on the other side if we uh Slash more than 32 U ethereums we could

end up in a situation that on a base level of ethereum there are risk introduced because now the amount of validat are not the same anymore and even ethereum has at the moment roughly 1 million Val when you look at the statistics you will basically realized that 95% of those validators are coming from from Big operators big node operators and out of those 1,500 the top 10 are basically having the network so basically a company like coinbase Cloud binance corus one pigment and so on have a immense power and if they are making a mistake this can have unpresented uh consequences for the

whole ecosystem this is one side like what can have happen for the whole ecosystem it can really blow up um I'm not even talking about I L if there is some smart contract Leach happening what this could lead to or when I lir introduces the insurance of attributed security which basically makes them their base liquidity uh layer more volatile because you have insurances something happens you have to this in account and now you have AES there you have always as a validator the opportunity cost problem which validator whichs to validate what are

you getting uh do they provide two sided token oneid token how many how much dependency does this AVS has to other rs's like if one RS is slash are we automatically in a casket where five rses are not working prop ly all these elements come together and with that the complexity is magnified to next like to the dimension unseen here and the here is the issue hereby is and the main issue is really that amount of liquidity which is going in I lay we are moving so fast we are over 10 billion there and I can really foresee I attracting something like 100 billion in in

liquidity and with that the responsibility on one side but also the risks on the other side of the whole ecosystems are much bigger much higher than the current def ecosystem that's the whole game can you help us understand really the um the risk that is introduced by uh the staking providers so you mentioned like chorus ones figments Etc um where can things go wrong at this level yeah I mean the there two elements the one is like the taking provider has their slashing mechanisms in place but there are this ethereum has slashing mechanisms in place but these

slashing uh events are not that frequent they are not happening as often like the most important metric to assess them is actually the up time uh of of those vals this is the one thing the second thing is like when you're looking at node operators when you go in the direction of R taking what this actually introduce is concentration coinbase cloud has 138,000 of validators corus one has something like I believe like 40,000 or 30,000 validators this basically means all rewards from risk taking are going to go into their Direction which will actually introduce more centralization for the overall met a smaller validator will not get this rewards he's not able

to expand the infrastructure he's not able to specialize his infrastructure to get more rewards on the side so basic is centralized the node infrastructure we have of course always the slashing mechanism of ethereum they have to take care of the validators which is working but this is quite good so far working you have the after time which is the base quality metric you want always to take in account and the only risk they are having at the moment from the risk taking spaces when they are providing their liquidity in the IP smart contract this slashing mechanisms if they are designed in a false way the logic is not correct or if something happens which we didn't see that this ethereum can be

slashed and huge amount of ethereum can be lost and you can really imagine it imagine a very popular AVS it's the new a of AVS uh it attracts billions in security and the logic of this RS is maybe flawed or something is not wrong not correct in the slashing uh procedure and then you can out of nowhere lose billions of ethereum this is the most extreme case if these node operators are not doing any kind of risk management a best analogy would be like you you have to see these node operators as almost like a crypto asset managers like they have to allocate their ethereum somewhere and they have to take a portfolio related

view where to put that what are the opportunity costs how much consequences this can introduce to them and how much how much exposure do they have to other node operator other OBS and so on so this would be like the node operator side so node operators will have to effectively create a portfolio of avss that they're validating for and act as I mean they sort of act as like you said this like digital asset manager um where they sit in between the protocol the ABS and uh the the the folks that are you know delegating or E to them at the at the ABS level the risk is primarily a smart contract risk correct oh uh

actually it's not not only smart contracts like AES uh AVS have a different logic like the thing is like you have the security risk of the AVS that's sure but you have also to understand basically the stability of the token like you are getting reward as a node operator and these AVS are not able to pay in usdc USD usually these are projects put trffic security they will pay you in ethereum and they are token and you have to understand this quality of this token does it make sense the stability is there kind of staking mechanism in place of This Ss that you ensure the stability of the token is ensured and the second part is basically you want to understand the underlying Lo of thiss how much security did they buy

how what is the cost of corruption what is the cost of profiting taking care of that all these elements are playing a major role on the other side and you as an not operator want to know this information if they have security mechanism which is wrong you taking risk if they have a token which is quotess you taking the risk so basically toonomic doesn't matter if they have a governance where one guy is taking control you are taking the risk and if you have a token which is very volatile you cannot be ensure that in 3 days from now the token will be paid out in this reward what you want to a um and then you have of course all the security related uh elements in place and the issue is like I is building this way that you have to wait 7 days is if you

want to pull out and you see now the logic why it really matters where you put your security into how do you think this is going to shape defi in the future what what does this um cause in terms of an impact on the on the future of defi that basically uh RIS taking is now it sh on a very good thing for defi because it brings on like taking braks like up to 100 billion of uh liquidity on chain and this liquidity goes into risk taking and from risk taking into defi some of goes in defi so basically defi gets more liquidity so basically

the rewards the speculation everything is now in place those Protocols are Al that's the shortterm profit they are getting the longterm consequences they are having is they are now dependent on the risk experence on the RSS on the lrts because now you have an issue you have the liquidity and you are having a stable coin if you want to call it like this which depends on the security of this RS and if this AVS slashes something this stable coin can be deped and this has done a huge impact on you as a protocol you kind of can almost mimic the teruna C like dpeg people getting panicked losing con

confidence everyone's throwing off uh and then you get the waves all over the space and through that like Risk Management is really crucial and risk management in a real time way will play a major role I I don't think like consultancy based models can can work on this side this is the first consequence so basically we are moving this real time Paradigm what we are offering and the second part um is actually these def5 protocols understand that that all this stuff is getting so complicated there is no way they can manage and they don't want to put that everything to one company they are now readjusting their protocols they're building this layered modular way of protocols where they are

basically isolating the liquidity from the control risk management and giving control risk management to various providers everyone can build their uh model um and then you can decide do you want to have a realtime risk management verified model do want to have a cond based model as a user and through that you get actual more stability into def space so defi is reshaped to that defi has to reshape due to the complexity of the whole Tex to add there another thesis we have in that sense is that the relevance of risk management overall will increase because staking and reaking is much closer to institutions and professional investors than yield

farming in Define so um we see already now already today Major Institution doing staking The Logical next St uh step is to go into reaking at a certain point in time and as Mario has mentioned also liquidity is much deeper already in in this space so it opens up um yeah a new set of actors and these actors are most likely more professionalized less Deion likee and in that sense then always comes um also risk managements into play yeah I think I think that's sort of the next step for defi is for the entire space to become much more

institutionalized and and institutional friendly um there's sort of a middle ground between full D gen and full trafi and I think that uh that adding more risk management and the ability to quantify risk um is is one aspect there's obviously you know all the sort of like Regulatory and compliance aspects as well um that regulations like Mika for example are um are helping bring us closer towards um but yeah I I think that with reaking and and particularly like the efforts of you know large institutional staking providers like the ones we

mentioned more institutions are coming into crypto and I think generally that's a good thing so what's the road map and um how can people get involved in ultrascope like can we already sort of create models is is the um is a product um available for people to use and what's the best way to get involved our platform is live you can already go on app. alc.org and build your own first Frameworks we started now publishing every day models people can use as an inspiration to build their own models for the space we have dashboards

and uh even a research tool available that you can get better understanding of the protocol risk you interacting with we plan to launch our test Network this summer and Main Network launch going to happen end of the year um as the co-processor side is still in development we have to do proper testing security and and on and so on before we are launching the coprocess especially if you want to get it like cost effective and automated as much as possible great and uh where can people find you

guys the best way is simply to to go on our website um that's that's our like Central Hub um what's the webite and in terms yeah the website is um alc.org and there you you can directly access um our platform it's it's without sign up so you just click on the website and can start looking around start doing analysis of protocols Mario mentioned already the research tools so you can start comparing metrics and also build your first models I would also add follow us on LinkedIn Twitter and we have a telegram

community group join us for that where our research team is now constantly providing updates how to utilize our risk engine great thanks for coming in guys thank you Sebastian thank you Sebastian for having us