Exploring the Future of DeFi | ETHDam 2023

foreign because now we're going to be discussing uh the future of defy opportunity challenges and Challengers so um I would like to also welcome to the stage uh gasparat and also seraphim welcome welcome no worries um so Anton we already know who you are but for the other two please could you shortly introduce yourself and what area of defy you most care about hey can you hear me yeah cool um Sarah FM from Lido um I call myself D5 expansionist I came up with that so uh that's what I do I care about it I care about liquid staking I suppose indeed and I care about making Diva efficient cool really simple and quick um so we started at a high level and um obviously there's a D and D Phi that stands for decentralized um and the idea is decentralization can um make blockchains I guess censorship resistant but in D5 also with the increase in regulations kind of really how censorship resilient is defy and how can it continue to be so it's not decentralized mostly most Protocols are not decentralized they're handled by a bunch of multi-cigs and you could override it um but not all of them which is a good thing so there's plenty of examples where it works perfectly well like liquidity uniswab which is it's nice it coincides with the fact that they're quite big but the reality is most D5 is not decentralized and whoever Works in this space know that completely agree and to add to this answer even when it's not a military it's not necessarily decentralized and to build decentralization you need some kind of inertia that often is something that you into weight or build up in no longer the time frame than what we had currently so we have a lot of protocols even with online governance that are completely not decentralized at the moment can you hear me okay uh yeah I think that uh every time someone is trying to uh censor the D5 or entire layer one it's kind of challenge for uh layer 1 developers and they are trying to fix it on their end because previously there was no much talks about it and it was not an issue but recently everyone saw that it's it's possibly an issue which we could face in nearest Future and developers are trying to solve it to make it much more free to have a freedom here uh as far as I can see the regulated Defy is probably a kind of subset of G5 and any other kind of restriction regulation censorship is for subset of D5 we will always have superset which work permissionlessly but G5 is such a freedom which allows subset of its users and projects to self-restrict each other with kyc or anything else so it's it should not be restricted to have restrictions and building on top of that like you mentioned kyc and I think one of the terms is kind of institutional D5 where you have to kyc um what are your guys thoughts on on KY seeing D5 protocols yeah I don't like it obviously like there's kyc on the secondary level especially something that just not composable within D5 people try to come up with like um tokens that if you transfer them you need to be kyc and it just doesn't quite work I think it could work at the primary minting stage like tether or like um certain things like that that could work but not on the overall level I think the better approach is to create assets that are permissionless like steak teeth or whatnot and put those assets within the hands of people that need kyc like track fire entities and hedge funds and family offices while we don't need to do that as D5 we just create a product and they can kyc it you know agreed I think it's a question of layers as well and eventually if you if you want to build that protocol layer we should avoid uh having those kind of constraints and remain neutral as much as possible and eventually if you want to build a higher level products then they will need kyc at this level but not at a protocol level that we're building uh I have cool example I think recently uh one of French banks released on stable token but it's not a real stable token it's kind of Meme and people had a lot of fun in Twitter about it because it's like non uh it's non-composable with almost any D5 protocol and uh trying to use it in uni swap or compound or anywhere else would just break these protocols because it's not compatible with them it's not even compatible with erc20 standard uh and uh it's uh what's trying to do I'm not sure if it's competing with any other stable tokens it's kind of different thing for some users we don't know who and that's where regulated thing is not playing on the market it was designed for the intention was to participate in this G5 Market but what we see there participating in own subset of restricted something and where most of the protocols are not possible to use yeah that's fair the um so like onto one onto the challenges there's a lot of hacks in D5 there's no way of getting around that it's just a reality of D5 itself but ultimately it gives the industry a negative rep both from the regulatory perspective as well from the user's perspective if you get hacked you've you've lost your assets and that's that what are your thoughts on Hacks in the default space and kind of what do you think the solution is moving forward yeah I guess I knew I think I know a thing or two about hacks so uh now I think hex is like a big structural problem because we have to be right every day and North Koreans or whoever has to be righteous once and the question is can you really build a multi-trillion dollar industry around something where you have to be right absolutely every day and they have to be righteous ones and I think I don't think it could probably within the current framework like quite a lot of people discussing things like you can't just withdraw 100 million bucks from a protocol within one block there probably needs to be some sort of a speed bump settlement times and that's the problem it competes with the idea of permissionlessness right so the question is is it the multi-sig that does that maybe there's a better solution but my point is we'll probably have to come up with speed bumps and settlements and things like that for large transactions and the questions becomes how do we do it do we do it in a way that is like traffic or there's a better ingenious way um without that which is going to be like a like a torrid website thing where kind of it will still be used but it would not be as big as we want it to be we want to be like internet kind of level Innovation right so um yeah we might need to do you have an inkling of what it might be like would it be on the application layer or would it be on the protocol layer yeah I have no I I have no idea people brought up ideas but they're mostly like basic multi-6 stuff which basically slows everything it becomes like like you know old school Banks where grandmas approved transactions every two days like HSBC nowadays I just but I hope somebody comes up with it because otherwise we can't scale and we're going to be like oh we're going to be like track fight then what's the point I think the velocity of the phones getting out is definitely an important point but I would also say diversity of the value getting in is is to put into question in the sense that only in defy you have that amount of value that that goes in code that was deployed like not so long ago and I think the more we the more we mature as an ecosystem maybe the less we'll see this protocol getting all the phones at once I completely agree with the currently we have to be right every day and that's a big issue and I think maybe indeed with some more um infrastructure block on on those protocols we might be able to restrain uh like big acts to empty your protocol in one block that being said I think also with maturity people will maybe deploy be more skeptical about putting too much funds into one protocol and because we we did see big protocols not getting act ish in a longer period of time and those other protocols with the most value in it so I think it's it's also about that kind of not necessarily due diligence but letting time do things in terms of how much money is put into one contract at once because for sure risk is a public probability uh I won't say that there are some projects in G5 who do educate users about risks and uh yeah there are a number of different risks different types of risks in D5 and some projects they educate users about that so some users can estimate better do some of their earnings deserve this risk or not is it higher or lower uh I think those projects should get more attention more attraction to be integrated in user wallets to make it kind of more accessible for users and one more thing here is uh like the greatest feature of all the Defy is its composability how different projects could build on top of each other and it's also important to understand here that it's a compounding risks of different protocols usually and uh yeah I I see that situation is improving slightly with the time like each with each next year we see that more projects understand what exactly they need from audits what exactly they need from educating users about using their platforms and I believe that it's getting better and better with time so we see that number of hacks is growing uh amount of stolen assets is growing but that's because whole Defy is growing and I believe it's growing much faster than hacks are grown and to add something I think uh you're completely right on one thing is that information is lacking for the defined LP the defy investor um we we tend to appreciate a lot that uh things are transparent but the information is accessible but not necessarily readable and without this information you cannot build proper risk management so I think as as long as we don't have that we won't be able to have proper risk management framework and defy and we will be exposed to the wildness of the space as it is currently could you maybe talk a little bit more about what are the types of information that you would want to see around risk management well you know for for starter um when one protocol breaks uh you tend to see The Dominoes of it and realize oh okay though so they were actually deploying liquidity there they were they were actually exposed to that vocal so I think like the interconnection uh to understand the compounded risk that was mentioned uh already that is pretty much needed like when you deploy when you put on a pool you want to know what are the weakest links uh who are the owners like how can like every things that can make the system weak you need to be able to to see that and then to put probabilities on top of that and just build proper risk management I think it's great that we have all those interconnection in the file that's the money Lego that we love but also it is what is bringing the the risk there so we need to understand properly maybe I could just add there's a devil's advocate I feel like with risk management and defy it might give us full sense of confidence because just because we think oh this can actually very liquid it's not going to get manipulated people start building leverage like uh positions on top of it so if you have a lending protocol people are going to probably deploy into it including dexes and because they think they know what the risks are they're going to do that more and more whereas people don't hitch for the actual tail risk that cannot be Quantified which is smart contract risk right and um it just shows you how systemic the entire thing is and how risk management is just very hard to ascertain when it comes to Smart contract risk like again if you have a bunch of Dex money plugged into lending protocol or a stable coin asset and that thing gets hacked certainly there's no debt liquidity there's no Dex liquidity there's no liquidations or other lending protocols and you see how a 2008 like scenario just plays out in D5 while everybody thought they knew the risks because that particular risk is just impossible to figure out um and that's where I think probably it's more about coming up with something more structural that just tools that show you what the risk framework should be like it should be something on the some kind of a layer one maybe level or again uh an a structure on top of that you know yeah that's fair and I I think one of the things you just mentioned was the person that's responsible or the people that are responsible for that and in the context of the previous conversations that we've been having around especially alexei's case where the argument is it's a smart contract there is nobody and responsible kind of how do you see that in defy where especially in terms of like liability when something goes wrong in D5 ultimately is a smart contract is it fair for the ecosystem to actually hold those people who designed it responsible or yeah you mean should people go to jail for that contracts or what can I say I mean um let me think about it if I have a minute it's a tough question I think as as really as we have the responsibility of being transparent um then when something breaks it's necessary like it's a responsibility of the of the person that put the phones in it like they trusted it eventually if you're doing the best effort you know by definition you could have you cannot do more so I think as long as you're transparent and you're doing everything that you can then um then that's the responsibility of everyone as well to manage correctly like how you you do things it's easy to ape everywhere and then expect just to to earn head drops and liquidity mining Cushing I can add that some of the most popular protocols they uh has value few magnitudes higher than those uh projects teams can afford to refund or to ensure uh using insurance uh and uh I think users should be informed about what risks are on them and what should could be compensated what not because for example like money markets they have billions of assets inside of them tens of billions some of them and it's uh absolutely clear then that if any anyone would hack it and steal tens of billions of assets assets those companies and teams who build those protocols they were not possible it's not possible that they would refund it there is no way the only way what would happen they could launch new version try to fix it the someone could investigate and try to return stolen assets but for sure their team and the company behind this they they can't they can't solve it with their own responsibilities they can't just refund it at all they could probably take some depth for example they could issue new token and gave it away to everyone who lost assets and in future those people will be able to get it but it's kind not a common practice at all and uh yeah I I believe that users should be informed much better on what exactly the other risks yeah I don't think it necessarily covers it when you get this like disclaimer straight away that you yeah way too much actually this should be in terms of services but not much people usually read it yeah that's the problem yeah and I guess immutability generally is a tricky concept because some protocols just don't need to be changed much like uniswap works like a robot liquidy works like a robot Lido doesn't actually need to change too much and there's protocols like lending protocols where you kind of need to manage it it's like a competitive Prime brokerage at a bank you need to offer new products you need to increase decrease leverage you need to that's where you've got chaos labs and uh Gauntlet and guys like that so it really depends on the protocol uh uh but ultimately I think you just can't hitch for everything so as long as you're not trying to commit fraud of some sort you know it's like in track five you just make mistakes it's fine you know you shouldn't go to jail for that uh especially not for like a good piece of technology like yeah that's fair um in a few minutes I'm going to go to the slider if you haven't used slido you just go to slido and you put in this code here two three two three three two sorry my bad um and uh yeah please upvote and choose some of the questions um one of my next questions which is very open-ended which is what are the opportunities you see for D5 Builders moving forward yeah I think generally for the future the way that we kind of get it get bigger we need to make sure that people that hate us uh need us and what I mean by that is like everybody kind of hates us right but they need to really rely on us like Banks middle officers back offices that's automated by ethereum blockchain in some way it's much harder for them to kind of new curse right or Lobby against us and that's what I think guys as crore or ZK sync are trying to push or uh stake teeth if that ends up on a Goldman Sachs in JP Morgan's balance sheet and they start coming up with Structured Products around it pitching it to family offices they can't actively nuke us as well they might hate us but they still kind of need us because that monetary incentives are aligned so I think the point is I'm trying to make is defy we started to realize we need to get leverage and we're doing it internally which is great and we just need to make sure yeah they're people that hate us need us and that's how future is going to be brighter for Builders um I think we've seen the interest interesting piece of technology being released recently in terms of privacy you're thinking that you can do with it it's in you know it's it's always a very easy to deploy and you can build so many new things that you will not be able to do it before I think in terms of layer twos and scalability we are removing a computation constraints that are now allowing us to build much more crazy things and interesting things in terms of cryptography as well um so yeah I think this thing becomes available now and very much interested to see what the builders come with and what new things can we build now that we couldn't uh two years ago [Music] yeah I believe that there are like 10 more tens so more great ideas which are coming to be invented in coming years as uh someone said bearish Market bullish development I would agree with that on bear Market it's uh much more much more less news about crypto and they distract developers much less and uh yeah I believe that we will see much more new protocols which were never existing before like I don't know decentralized stop-loss orders or anything else much more on-chain solutions uh something new and uh yeah I believe hackathons is probably one of the best places where this new solutions could appear uh I'm also myself going to do something on this hackathon but kind of out of competition I believe we have some ideas which could be pretty useful for others kind of public good and I hope we will do this on this weekend cool sounds good um this is kind of one one question um a slash concern is with the increase in layer twos you mentioned layer twos there's potentially the the the effect of fragmentation of liquidity for D5 protocols Bridges being like the interim solution um and like my question is okay I'm a developer and I want to develop within the evm ecosystem I then have to start making choices am I going on main chain or mainnet am I going on an L2 then which L2 do I now pick do I go with with a zero knowledge like how do you guys think about that decision process I think I think I think um multi-chain and multi layers doesn't necessarily imply fragmentation of liquidity I think you can choose what are you using for which reason and eventually with you can actually build good ux without having to to switch wallets networks and that kind of thing so I think of it and we're just trying to see pieces of infrastructure that allow that uh we're building ourselves something on stock net and we're like managing to do that without imposing that as a constraint for the users and we're eager to see how we will have apps protocols working in a very seamless fashion leveraging the best of the Technologies making the right constraints and compromise there and that would not necessarily split the liquidity we don't have to deploy the same protocol on all networks and just hope that liquidity will come everywhere and use Bridges to Arbitrage that I think we can build a much more efficient system yeah I think what's cool I used to think that it would fragmented liquidity a lot but what happened is layer 2's and layer one started to differentiate themselves like everyone has a different vibe arbitrim is kind of like a I know DJ whale kind of vibe being B is like a non-western DJ and stuff it's just different kind of users and I think that's the nice thing about what happened within the year that every ever turn that one has a differentiator a different vibe that you should tap in so if you wanna like you should probably go and order them at once and you tap into different user bases which is what like I like wanted lighter to go to BNB because it's like we're not siphoning away any funds from ethereum maina it's just a different kind of users if you go to arbitrary from BNB like wombat did they just tapped into a different user base so everyone's got a different vibe you should just you know go for five being a technical term could be you know so yeah go for whatever you like uh for projects I think it's important to wear start like the very first day chain is usually important and they usually what from what I see they usually decide now from using clear tools like should they go with optimism or arbit room or zika Singh and uh after they launched after like couple of months if they feel that their work project is like getting some traction it's not that hard to have uh it working on multiple Chains It's not usually an issue for almost no one protocol to launch on multiple chains and be kind of multi-chain but they don't need to support any cross chain actually and regarding the liquidity fragmentation non-layer tools it's uh much safer I mean when you use layer tools your liquidity is bridged via standard bridge in this standard Bridge of layer 2 so it has the same security assumptions as whole layer to solution so whole layer 2 network is the same safe as its Bridge because it's native and this Liquidator looks like fragments yet but it's not far away so users who Bridge their liquidity among layer 2's and layer one they do not almost have no any other risks they if they decided to go with this layer 2 they are fine with this bridge and I mean this liquidity is fragmented but it's not far away so for projects it's much more important if they believe that small warriors with hundreds of bugs should be involved in their protocol uh I think it's true that those wallets are more active on layer 2's because transactions are cheaper because it's hard to pay few bucks if you are trying to deposit 100 bucks it's a huge fee but on ethereum yeah people are still capable to pay tens of bucks fees but it only makes sense if they're moving some more significant amounts like tens hundreds of thousands um yeah it definitely makes no sense to swap 1000 bucks and pay uh 20 bucks fees yeah so it depends on the protocol if you are working on something like huge for whales for their deposits it makes sense to run on layer one uh maybe like layer one first and if you're launching something which would engage more wallets with hundreds of bucks on them it could be something non-vale oriented probably yeah you should start with Layer Two yeah definitely makes sense um cool thank you so moving on to the the slider question so I'm going to start with the most upvoted to try and get some incentive of competition um the first one is from Goldman what do you think of decentralized insurance or D5 cover and can that help with protocol hacks and how could it be implemented yeah so I looked into this at some point the problem is again it's as easy to hack 20 million bucks as you as to hack 100 million bucks so it doesn't quite scale so other thing most D5 Insurance actually so defy Insurance within D5 uh it's fine but if you want to tap outside of it for the large capital sums of insurance like D5 is just uninsurable pretty much because of that steep like because the cost of hacking is the same pretty much so that's why I'm saying that with increased amount of capital there should be increased amount of difficulty of hacking right and only then we can be insurable outside of D5 um yeah when it comes to Insurance within D5 it's fine it just doesn't quite scale with this like like Sherlock can pay you 5 million bucks of tvl but the tvl numbers on the lending protocol are in billions like just doesn't quite end up you know add up you know yeah I agree I don't think users are ready to pay the price that uh nd5 Insurance should cost um maybe it will exist later but right now it's just way too wild to be insured in my opinion um but then it's it's a business that makes sense so when things will be more stable Maybe hmm uh from technical point of view any insurance should try to connect all of people who are trying to provide Insurance sell insurance of some protocol or some arbitrary risks and uh on other side it should connect a pool of people who want to buy insurance be insured and I see the there is still a number of possibilities to build such protocols it's just about connecting pull to pull it's possible to build peer-to-peer so like you can buy insurance from someone concrete but it also makes important that you will return or it will your insurance will be halted in some amount of time but if you connect people pull to pull and people are able to join those pools permissionlessly and yield they pay or receive or the yield they pay is uh should be determined dynamically so to incentivize more people sell this insurance or in situize people to buy more of this insurance if it's too cheap so yeah it's kind of Market market-based solution which uh makes it possible to do it in that centralized way so the decentralized solution here means that people are connected to people like in money markets people are borrowing from other people so if you deposit something and borrow something those asset is you deposit is used for others to borrow it so yeah yeah cool well the next question is um how do you think about the future of kyct I guess that's know your customer transaction if I'm incorrect whoever asks the question please shout um is kyct contradicting the decentralized vision while it is an inevitable Trend to involve institutions in defy what does that mean I think it goes kind of relates to that um at the beginning where we mentioned KY seeing D5 protocols the gkyc a transaction how does that work I think it's both Kay seeing the users and also KY team so knowing the transaction as well so it's identifying the entire value chain of that transaction uh yeah same as what I said before I think it should be done on the track fire level maybe uh but I wouldn't do it every step of the way it's just not composable yeah agreed I would say as it was as I said previously yeah D5 here is such a freedom which allows subset of its users and protocols to play with self-restrictions do you think things like verifiable credentials digital identities and soul bound tokens May provide some form of solution for kyc in the space that enables composability or do you think that's all like a pipe dream no this could work I mean as long as you can remain pseudonymous on a base level and you have like some sort of uh you know like when you go on Twitter and people add you and you say common friends that's where you judge if that person's real or not like it could work exactly the same way in the blockchain if I wanna if I want to have a deal with someone I want to know do they know people I know um so I don't mind that per se I think that's probably going to work out it's just as long as you can access the base layer with pseudonymity but yeah it's going to happen probably yeah yeah fair enough um so the next question is um I guess an opinion about perp decentralized trading on dexes there are any protocols like unidex rooting swaps via one inch or GMX doing that are these a challenge are they challenges to sexes centralized Exchange I did not get the question itself I guess what's your opinion about uh Perpetual trading on decentralized taxes okay um I think uh it should expand in coming years because as you may see on uh traditional markets the spot trading volume is a few magnitudes lower than uh derivatives Market and I believe that in decentralized way in the access we will have uh the same uh kind of ratio so but but currently we see that generative's market is uh much lower than spot Market indexes but yeah this means that it it will probably expand I don't see why it shouldn't so yeah probably uh derivatives on blockchain is the next thing which will be which will get a lot of traction in coming years yep yeah I just want to say I agreed like crazy everything else uh we have to make it efficient and then people will eventually move forward towards the decentralized solution I mean I don't use much of a decentralized centralized exchange anymore um you know either that time it was maybe a bit harder but now it's seamless experience works great have liquidity enough liquidity for me at least so yeah I don't see why it would not be the same with Spurs yeah I think what decentralized perps did especially GMX with their GOP they created the new asset class and that's the big advantage of a C5 like you can take GOP and do things with it leverage it up as well I think the source of yield from GOP is is pretty cool because it's not so like rewards driven it's like a non-chain casino like there's actual cash coming from people losing money because they over leverage but that's kind of how perp trading kind of works and I think that because you can take that GOP token and recycle it through different protocols that's something that C5 didn't achieve yet so um yeah I think that's pretty cool it's going to keep growing yeah and I get a question to the Audience by a show of hands like who uses dex's more than sexes okay that's around that's around maybe just over half I think I'm happy with that what about you yeah cool well thanks ever so much guys it's been a pleasure yeah no um yeah no comment cool well thanks a lot and please give a round of applause for this panel of these panelists thanks a lot