ZK Panel

Transcript

all right uh Marhaba thank you for coming today to our panel I'm Danielle from W Source we're Venture Builder we're doing uh Investments and also techto market and I will be moderating this panel I also made the content let me introduce the panelist before we start so we have an Allstar panel here tonight tonight this morning so please applaud we have Sandy CEO of scroll love we have Alex CEO of ziki syn we have Vincent CEO of ziky link I hope you don't mix we have Louie Chief product of starware please and we have already applauded Anish who is the CTO and chief science officer of Panter

let's kick it off so zik is our favorite subject as we all know ziki is uh spanning from zero knowledge proof which is exists for as I checked just a few minutes ago I thought it was 30 years but guys next year it's going to be 40 years so I challenge you all to have a party next year to celebrate the sickest together with shaffi and Sil M and everybody else so before we start I wanted to ask Anish who loves the educating part by the way we're missing two mics guys do we have extra mics team coming okay fine so in the meantime yes thank you so I wanted to ask Anish because when we met met and we met a few years ago we had the

discussion because this is one of the Technologies like Ai and others that exists more than 30 years right and it's still not mature so I was asking Anish I mean how come how come it's still not in use so Anish could you give us a little bit of a wrap up of how it all developed and where we stand today and from there we can take all the other subjects thank you apologies again for a bit late somehow in my calendar it was misplaced by 45 minutes I assume this to be at 12:45 not 12 we love you we're happy to have you so so the way I would describe this is like you know initially as an idea it is a paradigm shift and when it became applicable then there was a difference between an interactive one

and a non-interactive one really the non-interactive one really took off by like 2011 at that point in time we still were using some elliptical Primitives which made it very you know throughput was very limited and that has been a large amount of improvement uh you know fry stock guys are here so changed the whole game quite a bit then there were like a whole bunch of other involvements new improvements and the most recent ones include both folding as in like recursive schemes and uh lookup schemes right so effectively what has happened is like if you were to look at the size of the proofs historically uh the time it took for us to actually do things and the through put we have it has improved

quite a bit you know if you wear somebody who has been looking at it from like 2013 like the real implementation in the web 3 space is more or less 2016 2016 October give or take a few weeks right it's the first real implementation that any of the l1s could actually have till now if you look at it you see significant Improvement like I would say even though we are still there if you're being skeptical you could almost say our competition is the optimistic mechanisms versus ZK right so in in in that in that real question the answer would be if somebody were to provide us with an insurance that underwrites the risk for that and the cost of doing ZK so one of the things I

kind of missed here is like fear on any l1s there's an amount of cost involved in doing ZK because that's a bunch of computation typically if you were to think about it what you call multiscalar multiplications and F fast for transforms the number the for transforms there have been a lot of changes different schemes have completely moved ffts you know mve out other parts of it lookup tables are come in but overall you still have a cost the the real what I would call the real litmas test would be can we actually have the throughput and the cost almost in line with the insurance cost you have for optimistic mechanisms when that happens I think we are in prime time all right we'll come

back to this later when we touch on challenges now one of their uh premises of zase is scalability and I wanted to ask you uh Vincent if I may to give us a bit your position on scalability right now what is happening right now and what are you guys doing in the field yeah uh uh Zer proof especially Z to uh have made the promise to bring the scalability to ethereum and I think we're still very early today and what we're doing we're building a l Le uh which you know bring down further the cost of transaction and it is aggregation of liquidity layer and

across different layer tools we're trying to build an aggregator that could be very beneficial for certain types of application for example for Lending for trading purposes and and secur for us we also have another platform which is application specific uh z r up uh similar to what a written the the circuit and optimized some uh you know certain up Cod and also we have uh adop adopted the uh State de uh we learned this from uh Zing to and this could be uh this could bring down the uh cost then the scalability today we're seeing uh in general because we're

not building like a general purpose Layer Two so uh I think this is going still early and we expected see this to be improved uh in the coming months and quarters maybe years I don't know maybe yeah I'm I'm happy to know from you guys what you think about the the when we kind of could could go there and to really get the to be able to uh uh have these uh applications that can you know Bo a lot of users and be able to support the high performance applications thank you so the way I have structured a bit that we will go over the main premises and Main offerings of Z and then we'll do another

round where we will tackle the challenges which will be scalability cost and everything else so uh I would like to cover so we have done scalability so far the next uh item I wanted to cover today is identity management using Z case and uh Sandi has uh kindly accepted to give us her inputs on where we stand on identity thank you switch hello hello yes yes excellent yeah so um I think there's been a lot of experimentations with onchain identity management tools in the past and I don't think there's been any that has been particularly successful apart from maybe worldcoin and um I think there is a

future where like AI makes most people's work somewhat redundant and onchain identity becomes an important tool to kind of um to kind of like aggregate your credentials and um like delegate resources in an automated way so that we can be free from the burden of work and just like relax on the beach so to speak and um the way to get there uh there are there are scalability blockers in between like what Vincent was talking about and there are um kind of privacy or selective privacy features um kind of in in between and then there are kind of um a bunch of kind of moral issues and

ethical issues and how data could be stored in a safe way and um and and and also how the whole entire uh ux and the whole like you know asset management process can be moved on chain so I think that's a that that that is a potentially a future that we're moving towards and um and I think to get there there are multiple parts of the components and we're just at the very very beginning of um many Explorations in this direction can I add something to this please so the uh I I I just want to highlight two interesting areas of identity that we see being developed uh one is direct cooperations with governments where

governments actually embrace blockchain Tech and start putting identity credentials directly on chain this is happening on ZK sync with the cork ID project where working with the governance uh government of Buenos ciris uh where they have an nap and they they mint uh credentials like driving licenses and birth certificates and so on directly on ZK SN and you can access it from an app and they are acceptable in in in the city of Buenos Iris I have to ask you if it's something that Javier M has started and yeah the the like since mes is is is like very Innovative and libertarian is and and like uh a very uh you like they are talking to them like we the they are looking to expend it to all of Argentina and also

to Colombia and Mexico and and like larger region so that's one way we see it and we like this this is happened in the past also in Switzerland I'm pretty sure that uh here in Dubai the government is also very Progressive very open-minded to to such things but the other path is even more interesting there is uh a bunch of applications being developed called the ZK kyc the way it works is you install an app on your phone and you can tap your passport and uh there is an NFC cheap on on on most biometric passports and you can and these chaps can usually sign messages and they have a unique ID and certificate from the government and you

can generate like the the app can generate a Zer knowledge proof that you are a national of a given Country and this country is not on The Black List and you are like 18 years old etc etc like some some like facts about you and then provide the proof of basically your Humanity yourid identity and some facts about you without disclosing your actual identity that's amazing that's the the amazing part about it is that you don't have to ask your government for permission because the like the these biometric passports are around for like 20 years a lot of people have them and uh basically governments cannot like say no because you you just scan it and and you can use it for these things and I think this is one very interesting area

to explore like how we we can do permissionless Innovation on blockchains um diing into existing identity infrastructures um from the from from the real legal world did thank you so much I was very insightful so you Sandy and Alex are taking me directly to the next question which is privacy now the Privacy is kind of tricky because I always thought that privacy is a bit of a premise of ziki ziki piece because it it it's actually what you have just shared that it's MPC and all this was you only share what you want to share and you can retain what you don't want to retain right yet there are privacy protocols on zase and here we have the gentleman on our panel who can tell us about it so Anish tell us please why you

guys doing privacy on top and I need to ask you to cover also the latest on aex development thank you so I would put my hand up and say something uh that's kind of very relevant ZK on its own doesn't really provide you with privacy this all the set of people who are participants of the whole thing gives you the privacy construct so you have a privacy set in that sense so ZK just allows you to prove all the things that's been said it works if you have a nation state with 10 people you have one in 10 chances of actually figuring out who you are and if you are flying out and if if you one in 10 there's only possibly one person in an airport so your privacy is completely and utterly gone so the question to me

was why was I impressed in privacy you know in general the the general construct of governance in implies that we assume everybody has a right to express their opinion on who they are electing and that's a basis of democracy when we get on the blockchains we have a strange situation all the normal l1s allow all transactions to be public I don't think anybody would be very keen to walk around with your wallet which you know broadcast to everybody in every part of the world every transaction that you have done you know in your whole life so this to me was a real Challenge and uh this is the reason like there the second order effect to this I should describe this is this challenge of information Arbitrage if

you have the ability to collect I mean now in the age of llms this is really relevant it require three things one is machines now it's h100 A1 100s or the Nvidia gpus it wasn't the case before but that's where we are now two is the data it's like if you have public data then you know you have access to data and three is a bit of a brain meaning like somebody who has ability to convert that and the gradient for that has come down quite a bit like you know the latest set of models the involvement from humans to do feat feature engineering has come down quite a bit so what does it imply the set of people like the Google's Amazon the fangs in that sense they have a real information Arbitrage on you they know what you're

doing especially if your financial data is all public you know they have a much better chance of actually recognizing what your next move is ahead of who you are so to me it's an unfair Advantage place for all these people right so you know in in the whole principle of decentralization and providing everybody with a fair chance I think it's really fair that we look at the possibility of providing at least bound of privacy where you have complaints and you have this disclosures the same way that tra trade F has in dark pools to in know users of up3 that's what I would say thank you anything you can add on Aztec if you know the latest development there

yeah I mean like you know there have been a whole bunch of people who actually created uh in a privacy with ZK Aztec is one of them uh zash mono the whole bunch of them right it's just that the regulatory situation has actually made people take hard choices the hard choice is like can you create something which truly complies and uh even beyond that there's this construct of being able to recognize funds being tainted so if you have a mixer where you you know multi Shi a pool where you have something coming in and something going out and if you have a split T join which is like you can actually put some value and you can split it between people

making sure that you can actually keep track of the tainting it's a really hard task and then you have to really say you know what do you want to really do so some people made the choice that you know we really don't want to take any regulatory risk especially in certain countries like UK a very difficult one so we covered two sides of the coin you know privacy and uh governments embracing uh ZK uh now there's one more subject I wanted to cover today and that's non-financial products and I have chosen the Zog starware Louis to help us on this yeah so um when it comes to nonfinancial application you know like um the the main news case I see about

from VK and mostly you know scaling it is uh the ability for couple of things the one that is most interesting to me is is onchain gaming um and I always give this example of um there's a couple of reason why Onin gaming needs EK and the main one is the fact that the reason why those Builder wants to be a launching game is because if you watch Ready Player one of the movie there is something absolutely dystopian about it which is the whole premise of the movie that who's going to inherit this this everyone's playing a game all their life and the game and start the movie start with the the death of the creator of the game and there is a challenge in the game to know who the first one to solve

it will win the ownership of the company so basically the whole word belong to one person and we have to hope that it's a the right one the nice one's going to take it over and when you look at you talk to every um auton word guide and on chain gaming guide what want is they want to create games that are going to leave the sur survive the death of the creators that are autonomous in the sense that they do not longer no longer be belong to the creators but they belongs to every participant and so the reason why you need the key in this kind environment is that we hope for those game to be able to run on Regular People laptop I should be able to run my own

part of my game I should be able to run my own part of the universe and so this is like sort of the first non financial application I'm seeing I didn't needly an obvious one but where you touch at and one of the one of those that makes me the most excited now there is also about the unex primative to financialization such as things like storage proofs uh trustless oracles so on Stark net right now or on Stark Weare ecosystem there is this team called ODU that is trying to prove the story of if to be able to extract arbitrary data point from a trust l source and provide that trust relas to any other network associated with it um in the same line of thinking uh we're seeing things like

um um um I have a blank so but like this is sorry about that it's okay so there is a bunch of places where the key trustless is becoming big and thing that I'm very excited about the first step being obviously a new word and uh I think it's going to be the first start of a new type of internet internet where servers would just be running trustly thank you so we have covered some of the major topics and from now we going into the second round where we're going to be tackling the challenges so one of the challenges known with zase and actually it's a little bit related also to you Louie but the question goes to someone else the languages used so uh the

challenge is specific languages that need to be used and on top of it um how do we on board web two Developers to Z I would like to ask Alex to give us his view sure uh and we ZK is maturing as a industry and just like the Computing industry was maturing over decades starting with people writing programs in machine codes on you know like this cards with dotted uh holes in them which were fitted into these big frames and then from there to assembly and then from there to see and and like like this low level languages and then

progressing to higher level languages and then to to like specialized languages like JavaScript for for uh uh web applications um ultimately the ideal world is where you write in the language which is specialized and designed for the application domain you're building for like if it's smart contracts it should be a smart contract language and not something like ZK specific right today for for For Better or Worse the linga franka of smart contracts is solidity like maybe it will change solity is not the best language I honestly hate it um there are like better approaches like I I I like moves inspiration from rust I

like uh actually like you could write smart contracts in Rust itself like there there are ways to to do it better but for now solidity is the main language and so like for us it was no-brainer that we have to support solidity the moment we could do this uh on zkm we we did it the very first zkm LOD by znc uh like basically support solidity you can take any contract and just deploy it in the future we should be able to support any languages for building privacy applications as well as building smart contracts and I I I can predict by by the end of this year we'll see ZK infrastru structure being extended to anything like we like we

work on something like this risk zero is working on this sp1 is working on this like you will be able to just write rust code compile it and uh generate your knowledge proofs for it thank you Alex so another subject uh under challenges is fragmentation so what are we doing with fragmentation and what are the fragmentation standards and here I wanted to ask you Vincent if you can help us Yeah we actually have a little bit of experience with uh uh all the different L TOs and different languages uh with Cairo we have a few developers that have um contributed a little bit to the codes and also with Solana with rust and also with uh uh ZK stock and you

know fragmentation of these languages and know fragmentation of the technical standards and know so the developer toolings actually a big challenge for developers some applications of us want go in multi chain and their like uh stopped most of the time they're frightened and we were in conversation with a lot of developer teams that actually want to launch application for example uh using ZK stock and they want to run their o own chain maybe and maybe on stock too uh they have to make a choice so first of all they need to make a choice which they're they're prioritizing and this market and people are Russian and this this choice is difficult most of the time and uh and this create the fragmentation of liquidi

because you have to go one chain you got to create a liquidity pool there and then you need a faster Bridge or you need a you know especially if you go onto you go to op and it's hard for you to really to be able to cope with uh something at the same time on the Chain uh yeah the this is actually a very pressing issue that we're saying and people and the users are feeling the pain then the developers are feeling the pain too to have to make the choices and how to go to Mage in multiple times and cope with and with limited resources we we keep receiving interest you know to to to uh help them to uh deploy or uh you know to be able to solve these

problem for them that to come combin assets liquidities uh of course it's not always about the financial applications but if you're doing nonfinancial uh application probably less of a problem at the moment uh but from the users point of view you know having all those Chang and the fragmentation fragmented user experience is also painful uh yeah so they have to have gas tokens and so we need uh account abstraction and also we need a chain abstraction there are so many chains already and there's going to be more unfortunately and there are so many robs and probably will be more and so yeah we need to have a a unified standard somehow or we need to have interoperability but interoperability does not only mean you know Bridge asset over is uh I don't see how this is is

happening in near future so we're taking this approach somewhere in the middle try to uh bring this change together uh and we'll be in a unified layer right our layer three this is quite unique approach to be honest I don't think the market is really noticing this yet and you know be able to do this with zch proofs just trying to maintain the security trying to you know to still be able to settle and inherit the security from e l one one and we if if some application wants to go further they want to you know unify and with us other layer tools there could be a little bit more of trust assumption using a third party like decentralized nodes of

oracles or something to oversee the states Roots synchronization course different on Layer Two the beautiful part is that we're able to do this with SE layer one and then we be able to solve this problem but in general fragmentation is a real problem and that we need to solve I think all these layer tools are great amazing technology but they're building their own ecosystem right so they with they're growing fast we're seeing this challenge you know for users and some application Builders thank you very much Vinson so one more subject I wanted to cover is voting uh Sandy could you take the voting on you voting um yeah I think web three is quite unique in terms of different

experimentations of levels of voting but the running of blockchain is a very it requires a very specific knowledge and domain expertise so no one has really tried like direct voting and when it comes to like Security Council it's always kind of within a certain body of people who have been prev vetted and so I think there have been some experimentations with like delegates or stewards whatever you want to call it you can think of it as like a representative voting mechanism um but that only works for more generalizable topics um so I think um yeah so so the big decisions regarding like upgrades

design directions and research um I don't think that part is really possible to be like super decentralized and and and voted on at least for a very very long time um even in ethereum there's are so many different experimentations just like there are so many different experimentations of l2s um the the the the the main protocol design decisions um are still you know it's not centralized but it's not fully decentralized and it's not like voted upon uh just because the level of expertise required to participate in these conversations are are very very high but I think the ecosystem as a whole evolves to decide and the market as a whole decides um what takes traction and what takes force and I

think this is um very very unique to our ecosystem thank you L you want I would just want out to where where thek does help in system of voting because there is like the obvious question about direct democracy and and uh the the the lurgy of the average voter uh but there is also a reason why most of the unchain governance today is just non done on chain it's just Cost like it's just goddamn expensive like Aragon died out because he was a million dollar per vote a million a million gas per vote at the time in 2017 it was not that too expensive then it became too expens afterwards and so we're we're seeing at least in the context of snapshots in snapshot is a leading voting tool in the market uh trying to offload that the

tallying the ination of of of um of votes to the L2 and send back the proof to the ra to L1 and this is what snapshot as has been doing on Stark net and we and especially with alongside the trustless bridges trustless Oracle is talking about with erus I'm expecting to see a lot more of those tool not only on St everywhere to emerge and to become the standard wave for um managing governance even if it doesn't invol everyone at least manage the government itself and guarantee unchain building thank you well we have one big challenge left and I wonder if Alex wants to take it the costs can you um sh the costs of well take it to General like whatever

you um there are really good resources to compare are the costs of rollups today uh one is called grow the pie the other is called gas gas fees. there used to be l2b uh doino but it's it's like it's a bit out of sync and if you look at those the ZK rollups like ZK sync and Stark net are on top of the listic we we are cheaper than uh basically than than than some of the optimistic rollups uh right and so they and before ap4 chilling a little Shilling go ahead I mean that's just fact I mean that's just fact I think the question think that works the question

is how cheap is cheap enough like what is the is there a state where it's just like that's cheap enough or it's kind of an endless scaling process to zero I I think no like it depends on for like for on who who you ask for the end users there is some threshold under which psychologically it doesn't really matter you know like if payments are under 1 cent whether it's 01 or 001 it it doesn't really matter right for Traders massive difference between 0.1 and 0.001 massive difference right because they can do a lot more higher frequency transactions like if you think of high frequency trading with like thousands of

transactions per second it matters to them right so like you will always want to lower the costs and will always be in this race but for uh the like for me like from from my perspective the the mission of ziki sync is to bring Mass adoption to crypto to on board 1 billion people on blockchains and give them sovereignty specifically for from from this idea of freedom and empowering people to have the control of their assets their lives in their hands and they in and and increase their prosperity for more economic freedom and for from for that I think there is a thresold like under one cent it's going to do it like if we can scale it to 1 billion people and we will will be able to do it with validium and with the

ecosystem of like interconnected DK chains on DK sync and hopefully on other on other ecosystems as well so I I I agree with you 100% I would argue though that while the existing use case can be covered with one cent transaction and transfer and so on um to me and actually what I I'm bullish on and very very optimistic is on the abch front is actually free transaction and by free transaction I mean in Sovereign environment where the the app itself control the whole environment and know that they can make money or they can make Revenue outside the regular deed like dose protection that Fe mechanism is so for an exch for an example an exchange cannot leave when another book

has fees that is no matter how you turn it and no matter how you want to put it an exchange where I have to can pay to cancel a transaction cannot succeed and you've seen that all across crypto forever and funny enough starware specifically has like quite an amazing pmf on pers with starx and all our hisory around the project building perpet exchange on top of a St stack is the reason why they would choose starkware and starware stack is because they can do free transaction actually don't usually use our brand because that's not what they are trying to sell to their users they're trying to provide self custodian trading and the only place they can get that is gas gas free but where I disagree with tiny bit on

the apin thesis is the uex of it is nightmarish and and and people don't think so much about it but the imagine you have like a million app chain tomorrow which would be comparable to having a million server tomorrow which is like quite normal like we have a many website in Internet like probably like a billion which probably is relating to uh 100 million servers worldwide maybe like a million servers I don't know how many Serv in the world but probably a lot um if my discovery of my wife it requires me to go to every one of these chain and servers block exper to understand how much money I have we are in a freaking mess and so the current format of vchain as we know it today is something I'm

don't believe in I don't believe in the conduit model I don't believe in launching your app chain the way we are doing it right now which is completely a completely independent infrastructure and um actually we're going to present soon and work something I worked on on a new design where there is a lot more flexibility on the way your prent assets and asset bridging and how you this operation can work in the long term I'm looking at you to to see how much do I have left because we want to do one last round five minutes okay excellent so um as you have heard so far there are many ways to get to Rome everyone here represents a different way but all together they agree on how it works so

um one last round maybe guys what are your specific challenges right now where do will you focus this year I mean try to make it sweet and short sure so I I have to follow up something Lis said before because that's part of the talents have as well so he's alluding to this construct of utx versus accounting so what effectively he's describing in my head is like okay having an accounting mechanism which is globally visible and that applies for privacy as well so for me the challenge is something like what he described can I have a situation where the cost of actually doing Discovery so typical privacy mechanisms are utxo based so essentially there's a workload

associated with how to find out how much whatever you have and the other part of it is like will The Regulators of the world understand the value will they see parity in actually having the same level of privacy that we now have like in Europe that's very clear description of what is you know what you know how much information you allowed to give and you have the right to be you know forgot on like a whole bunch of things and when they come to the web three World they seem to completely forget the fact that all the things they said like literally five minutes ago doesn't seem to really hold for their argument about F3 so I would love for that to be valid and as everybody in this panel would probably be interested in having you know cost

and throughput and all those things get into a steady state where you know you could almost get to a place where you could have any transaction and any transaction really makes sense thank you so in term of challenges um there is a lot of we are in the bull market right it's nothing new here uh we all know that's why maybe maybe I mean I think we are are we are we guess I guess we are and um I guess for all of our project the there is a lot of noise and a lot of very short-term um move to be like you know like when when a builder wants to build an appp right now the diversity of option depending on that timeline is gigantic and um some and bull market tends to favor shortterm Focus which for

most of our application I will arue here have the advantage of being more complex let's be honest it's more complex to build on any other system than let's say going on blast and so the biggest challenge for me at least is to make to to convince a builder that you know you maybe it's already maybe like that they should be building on a longterm time frame than they than they are currently right now thanks sand um so my mom asked me what can I do right now on chain that's fun and my challenge for this year is to try and answer that question wow Alex well we're decentralizing zik sing that's the

priority amazing we love it yeah our our challenge is uh I totally agree with Lou to to you know deal with the short-term choice that we were forced to do in this pool Market if we are and and a long-term thing and our focus is right now is to convince and this top tier developer community that this is good choice so this this solution can actually work longterm we're still in that process amazing thank you so uh what we tried to do today was um on one hand to educate the audience and give you as much information as possible on different approaches and what ZK can solve on the other hand we wanted also to highlight the current challenges and

opportunities so I hope we have done this I think our amazing speakers Allstar panel thank you guys 40 years zp I challenge you next year thank you Daniel thank you questions can we can we take some questions do we have some questions wow uh thank you for the great insights my question is mainly to Alex um um I I just want to understand like how confident we are on the multi-layer approach of scalability like L2 it's like regardless if it's optimistic rollup or zero knowledge rollup and L3 compared to monolithic approach which we see on

other chains like Solana it's like one is like um One One is using um it's it's like you compare or saying that you need to fix the internet speed at 1 megabit per second and then you need to iterate on the compression algorithms without like scaling it so monolithic blockchains they are like creating like parall execution just to like to to scale the chain where like the multi-layer approach with L2 is like trying to um like mer merizing the data and then like create more more performance and more scalability I just see I just see like two different narratives on the space on the horizon

and I I'm I'm kind of like from a user perspective confused like which which one is the direction great question and I I I have a very simple answer for this everything that alternative Lans can do we can do on ler toos but not the other way around if think think about like any property that they can bring there paral like optimistic parallelism sure we can do it on the rollups um like faster execution faster confirmations like very fast consensus mechanism sure we can do it on rollups with the uh sequencers decentralized sequencers in L2 like the Z SN is working on this other teams are working on this everything that is

possible like new virtual Machines of course we can plug in new virtual machines but you cannot have the properties that ethereum has which rollups inherit this ultimate resilience and ultimate decentralization and easily put them over on the alternative l1s and that's a difference uh I would even give a specific case of this um you know let's take Aptos andu what their main Innovation is BL STM it's coming to production on Stark net in two in three months so like the the you can and that's maybe I think Alex would agree with me you can retrofit any sequencing Innovation yes on a ZK rollup the retrofitting a zik system on top of uh

existing Tech is an insanee behavior and every time reading a deck they say I'm going to prove Sol us ra zero I'm like good luck yeah may I make a tiny plug in related today it's 3M we're going to have here a clipse that are doing salana and we're going to have pontum that are doing move so I challenge you guys to join whoever can and the audience to come back on this one because we're going to Alex apparently not available but would be really nice discussion go ahead I just want to say one thing which is like it's hard as like you know it's like

right now in AI you know the biggest issue is not so much like there's gpus but even finding the guys who know how to make the right model that works is hard ZK is the same thing there is very few people in this planet that can make the Innovation required to get to the speed that you want to sustain the throughput let's say s and I'm always skeptical that's what all I have to say so far of the new come back at 3:00 I need your input so I'm going to put my hand up and say something so I'm full disclosure I worked on two of the top five l1s I worked on a scheme which is a ZK scheme my team designed a whole zik protocol worked on AI sure worked on ethics f x risk done like ridiculous amounts of work in all these things I

see all the comments from everybody I agree at a high level what he's saying is true given anything you know it requires an amount of training an amount of intellect an amount of interest it's a product of probabilities smaller probabilities it has a power lad distribution the total number of people in the world who could actually build a really what I call scalable not decentralized distributed system you could probably count and probably in you two fingers two hands sorry okay and maybe in your toes at best I've actually interviewed one of these guys for one of the L1 okay going back a good few years ago they tried to move them from what you call a web 2 company a Princeton PhD a web worked in

a web2 company so all these guys are there's a limited set of people you can deploy it the way you want it's the market who kind of markets and the VCS who kind of dictate how this blame is going to be played if the VCS were to come out and say tomorrow you know if you use any of this Hypes we won't give you money changes if the market really says this is going to be really successful absolutely the game Chang so this is the game being played for two sets of players that's how we see it scarcity thank you we have one more question I see yeah very quick question actually so I was interested if you can elaborate more any of you guys actually that don't care who answers this but like the liquidity fragmentation problem right and how zuka could actually help

solve this or any other technology like in which direction we would be moving uh because yeah we all know it's like a pain if you want to swap something on one chain then you have to Bridge and wait and maybe you don't even have a direct to that chain so you have to like have an intermediate or whatever how could we unify this into like one UniFi pool or something like that I'm I'm going to start if you loow me um so I love thing the market calls liquid fragmentation I like to call that state fragmentation because that's what it is it's a question of State like where do I store things and the the so there is two aspects of this question is and then going back to the Discovery problem so the for me the bridging itself is not really the issue we can find we can do

Bridges the real issue is how do I even I forget I have this asset somewhere this is like weird I I have many I have like okay this is on record I lost a nonzero amount of money on beatrex because they went bankrupt and I forgot I had like a non-zero amount of money there like it's just a fact of life and and that you will not see things and so um the the fragmentation problem come from the the It Is worsened by by optimistic approach because there is no way to sort of like create this sharding into a chain VK you and I'm going to as I say like I'm hinting to something I'm working on to like a form of ZK sharding where the

core information can remain on the L to itself like the money how much money how much how you you own how much asset you actually own on the chain and at the same time offloading the computation to to to an execution chart and how you do that is very simple I mean you you would just have a nap that is running your own thing and they would just take your transaction batch in on their own and apply the state the money is still on there too they're still running there so this fragmentation is not so much there is two fragmentation there there is a state fragmentation and there is a synchronicity fragmentation can I do synchronous synchronous stuff between various app you can I mean to up to limit there is always like limit to how much synchronicity you can get but on the a synchronous approach Discovery is

s fragmentation is mostly about synchron about Discovery in my opinion and that's going to be solved very soon by ZK V do you want to add something um actually I think uh we all have like different uh approaches to solve uh liquid liquidity fragmentation because this is one of the most painful problem right now for users anyone that is capable to add value here is able to capture a lot of value and you know and attract a lot of users for sure and there's hyper bridge and there's also uh stagnant as a Madara to solve this problem right and also there's a op

super chain and a polygon cdk people trying to solve this problem now with the different Tech stacks and so we're a little bit like a more neutral so we want to to offer something that is Tech stock agnostic to be able to you know abstract across these different TCH Stacks uh well the end game for hybd change maybe cic also probably will be able I don't know how that works but I need to learn more and the problem is that the we still not able to solve this to have synchronized uh uh composable transactions of course different change this is not something that we see happening or we don't find a solution for that what we're able to do

is to have a a layer that c can aggregate as much as possible and you know uh but this also depends on how much liquidity you could attract is actually not from the tech side is is I think this is still limited uh we we only see this uh the best trade-off that we can find that we at Zing can find at the moment is to build this unified layer to be able to tap to different layer tools but the the the the challenge for that is the finality it's not fast enough so the user experience is still not great enough but if our c is that in the future if all the chains

are ZK ZK chains and all you know approved and and then finality is a lot faster than what it is today then we could have a great user or and a lot of better user experience then you could have you know enter funds and exess lot faster and settle those transactions and is a lot more friendly and secure for tradings Thank you very much I see is holding his passport so I get a sense of urgency here thank you so much guys for coming it was an amazing panel thank you for the passion I think the audience felt it