L7 - Exploring ZK Network | by Daniel Lumi | BlocSoc Lecture Series

Transcript

um I guess we can get started so it's nice to meet you all I'm excited to be talking to you today uh my name is Daniel Lumi I'm the senior product manager of ZK stack uh for ZK sync and today I'll be presenting to you guys rise of the ZK Network and ZK Network and ZK stack are our vision for scaling blockchain to to the mass markets to actually be able to scale where everyone in the world can use it and just a little bit of legal disclaimer that um we have a worldclass team of cryptographers of Engineers of mathematicians so everything we we talk about we think is factually true but we

always encourage people to do their own research uh don't trust verify so first we'll talk a little bit about what is matter Labs which is where I work uh and how does it relate to ZK sync so matter Labs is one of the leading zero knowledge proof companies out there we raised over $250 million and we really have a worldclass team of again Engineers uh cryptographers mathematicians uh from companies like like SpaceX Apple Google um and we really like our team really is is um quite excellent uh we're the leading uh Z blockchain out there so not just in l2s but in l3s um we have our blockchains

proven with over 30 million monthly transactions uh currently each um ZK chain so ZK sync EUR for example can handle over 200 TPS it's about 150 to 200 depending on the transaction complexity so uh deck swaps would be closer to that 10050 um that's compared to some other ZK blockchains like polygon which handles three TPS Max um uh but all of the l2s they they can handle essentially between 3 TPS to to uh 2030 real TPS like deck swaps uh but we perform regular upgrades on this so we should be seeing 500 TPS soon and then later on to um a th um and Beyond uh as an L2 we're

secured by um ethereum the full security of ethereum uh and we use Z knowledge proofs to to achieve that and our goal for ZK sync is to make it limitlessly horizontally scalable so you can get it to 7even billion people and Beyond while still being cheap at any arbitrary scale and keeping all the security and decentralization because that part is essential like we can't make compromises if we start making compromises there's no point in using blockchain anyways you should just use AWS or the current banking system right so again we are uh Layer Two so we do offchain computation off of ethereum

and then we use knowledge proofs to essentially um uh to verify to ethereum that every state transition happened correctly and all of those thousands or hundreds of thousands of transactions um essentially were executed correctly and in theorum does not need to reexecute those transactions so we're scaling ethereum using zeron knowledge proofs and first we'll talk a little bit about the Dilemma as we like to call it before you jump on me for the trilemma The Dilemma um in this is sort of the problem that every single blockchain in the world has it's a feature but it's also a bug um and this is true for every blockchain until this new vision that

we're building out uh which which will become the next generation of blockchains so just to get an understanding of why does blockchain not scale we'll take a look at first of how the internet scals and just as a mental model let's say you have a video game on your comp computer and then you boot up Excel and then you boot up Chrome at some point your computer is going to run out of resources like that Excel is probably going to crash it but the reality is like you you cannot run every single application in the world on one computer no matter how beefy even if you have a giant gaming PC or super computer it cannot handle every single application being run in the

world and so how does the internet scale or fix this problem how do they deal with this problem well let's say that you have a thousand people wanting to pre-order the new iPhone right so the first thousand people get it from server a um and that's fine but if you start having hundreds of thousands or millions of users at some point that server like in the previous example is going to run out of resource it just cannot handle all the throughput in the world so what Apple does with this or any website any web to the internet this is how the internet scills is they they copy the state of that website onto a second server and then the next thousand people or whatever amount of people can get the

website from the second server or third or fourth and you can just sort of keep scaling and the user never has to be aware of which server they're using um so the internet essentially scales linear to the amount of resource allocated to it so if you have one server with X resources you have One X scalability if you have 10 servers with the same resources you have 10 times scalability or 100 times or a million times as long as you just keep adding more servers to it and so how does this compared to blockchain like why can't blockchain scale and this is the sort of fundamental feature but problem in blockchains is we need to reach

consensus so every single node on that Network needs to reexecute every single one of those transactions that the blockchain handles so again in this case it's like the Apple example but in this case no server does separate work from each other they all have to do the same work over and over again uh 400,000 times in ethereum's case um and this is a core property to why we use blockchain in the first place because you're trying to create this sort of global verifiable truth without actually having trust so that re-execution is essential to it um and but the problem in blockchains again is you cannot have every single computer to reexecute everything like

blockchains don't scale anyone that tells you otherwise like some of these other block alt L ones uh are saying that they can scale are either lying to you or unaware or just it's marketing there's a lot of marketing uh in this market like this should be really really obvious no matter what fancy words they use you you can scale blockchain period right so blockchains have this sort of issue where they need to choose between either you make it very scalable by increasing the node requirements you're going to start having beefier and beefier uh servers like salana does and it can only run in very very few data centers in the world which essentially

defeats the whole purpose of blockchain there's some point where for example salana cannot keep scaling even they don't pretend like they can scale forever they say they can scale at More's law which if you know about it's 2x every 1.5 years that's a really miserably slow rate of SC scaling and it's going to take us forever to get to mass Market if that's really what we're aiming for or you can keep the requirements or the scalability very low but then you have Maxim decentralized more and more nodes creating this one Global uh layer of truth right that's something that like Bitcoin and ethereum optimize for and as a layer two we can benefit off of that while still increasing scalability and I'll get like this is what we'll talk about of how we

can achieve that so again blockchains scale differently than than the internet so if you have one server with X resources you have One X scalability in a blockchain but the problem is if you have 10 x servers or 10 servers with the same uh scale build or the same throughput you have still One X scaleability if you have a million servers on this network you still only have one X scalability which in ethereum's case is like 30 transactions per second right so it doesn't scale with the resources that are allocated to the network it it it just gets more decent caliz more secure and so what about if we just use cross chain like what if we just keep copying ethereum making new versions or do ethereum to

salana please don't do cross chain please please don't do it it's really unsafe like that that layer of trust in between will never be as secure as either one of the blockchains that it's connected to and even worse than that it becomes a a one like if one of the networks or uh on the the entire connection list that that cross chain uh Bridge connects to uh is manipulated or hacked um or goes down you can essentially steal all the money and all the the different blockchains so it creates this extremely weak point that again completely removes the points of using blockchain in the first place so you can scale that way okay so blockchain don't scale it's

all over I should quit my job go work at a bank screw it we tried we're sorry we're sorry metallic we really tried don't worry this is something that we were working on from the beginning and we call this the ZK Network CK is the end game I hopefully I'll convince you by the end of this um so the ZK network is made up of two different parts the core and the framework and the core is actually what we use for ZK sync era so if you've heard of ZK sync L2 it's era and it's the first ZK chain that is on this uh shared Bridge Network and the nice thing is like any

of the innovations that we make for era you can actually just keep copying onto new instances that are all based on this shared Bridge I'll keep coming back to this but any one of these hyper chains or these ZK chains get the same benefits that we've been building in Era so they become shards that all have the same uh features with in the future with something we're developing called CK IP those shards can be actually completely different so you could run salana VM on this ecosystem or move VM or completely custom implementation that is not a full VM even uh and it can still be interop interoperable in this connected ecosystem so those core features um

we've been building out over five years here's some of them and we'll go over the this list of core features so again we recognized this problem of like if if you pretend that 30 transactions per second or even a thousand transaction or 10,000 transactions per second can get you anywhere you're lying to yourself like that's never going to scale to mass market adoption right so currently each of our chains again have uh 200 TPS much more we have incoming like we can get to a, 10,000 per Shard right and this is not as simple as it seems you need to have a highly op optimized virtual machine you need to have a highly optimized databasing system the network

system this took us many many years to get to the point that we like we have the most scalable blockchain and you'll understand that in a second so for example like we have lower fees it's on rollup mode it's about three cents of transaction right now that's going to drop soon even more and with validium we're going to actually have lower fees than just about any alt L1 especially if you understand how fees Act actually work in most Al L ones so salana transactions for example they cost N 9 cents to 19 cents a transaction but they essentially artificially inflate the token and just dump it on the community creating people needing to essentially buy a lot just to keep the price stable uh it's about what was it8 billion a

year or something that they dump on normal people to artificially subsidize the transaction fees so these alel ones don't really have the fees that they pretend to have in our case that three cents is actually the cost of operating this network like that's the real fee with validium it'll really be a fraction of a penny and we'll come back to why validium is even better because it'll stay a fraction of Penny forever in fact it'll just keep dropping um in these protocols again we've hyper optimized also the prover of people think that ZK or zeron knowledge proofs are the bottleneck in these blockchains they're really not we've we've hyper optimized the proof system to the point that you

can run it on essentially consumer grade Hardware at least consumer grade gaming Hardware so like my MacBook can actually run a fullprof for a network because it has more than six gigabytes of GPU Ram like that's really little that this is we' our team has done an incredible job of optimizing it to that point um and one of the reasons that we have uh one of the lowest fees um is a decision that we made uh early on which is called State diffs so instead of publishing every single transa or every piece of every transaction we essentially publish the Delta the difference between one state transition to the next so for example if I send someone uh in the school one e they send

it back to me then I send it back to them they send it back to me and this all happens in one batch we don't update L1 at all because that doesn't matter to security at all like in the end I still only have one e and and whoever I sent it to doesn't have that one e so you don't need to actually publish anything to L1 to retain the security like we still keep that data because your transaction history is important if you're doing taxes or seeing what you what you've done but that doesn't remove anything from security and this sounds like a small decision and it's the state diffs are confusing to people but the reality is it's actually a massive effect especially for us getting like real trafi applications or real world applications so for example it'll make

any transaction cheaper in the first place it'll allow you to have like stochastic price updates so like in a Forex Market or in a stock market they don't update the uh the Oracle pricing every couple minutes because it's too expensive on ZK sync you can literally update the Oracle price thousands of times a second and you don't actually pay you only pay for one storage slot so you only pay for once for saving that price difference um which is really remarkably different than any other blockchain enables extremely high frequency quency trading because you can make an order you can cancel you can change the price you can you can keep like and like market makers this really matters in traditional Finance this

super frequency really really matters and you can't do this on any other blockchain while only paying one storage right uh it allows much more complex transactions it allows reoccurring Bank payments which people and Banks needed in their normal life right it allows um a system of onchain privacy which we'll get back to in a bit um it's not possible to really have offchain data in optimistic l2s it's really not possible uh not securely or make at that point you become a side chain and if you have offchain data you can essentially have a private chain and like privacy has become this sort of bad term but the reality is no business will ever Deploy

on blockchain no bank will ever give away all of their customer data the Enterprises have Trade Secrets they have customers that they they don't want to put everything on chain if everything's public we're never going to reach normal businesses um so this is essential actually to be able to onboard all of traditional Finance as well as just normal companies like apple Apple's not going to put their customer data on chain uh if you do nft mins or like you have your identity as an nft on a blockchain you can do like massive batch mints of nfts for and pay one storage slot it's it's much cheaper on our blockchain and identity is actually really important because we're already

working with um cork ID who's the official digital identity solution for Argentina so right now it's it's a choice whether you want a digital identity but it'll be forced um like it'll be essentially you'll have to get a digital identity soon in Argentina which means every single person in Argentina will have their identity on ZK sync from there you get a lot of more like all the payments will start going because you can't have cross chain safely like we're going to onboard the entire country of Argentina just because the like besides the fact we have scalable technology that nft minting actually matters um if you guys have played around with blockchain you'll figure out that it's it's quite clunky the ux it's

not great there's a lot of signing where like we kind of like the internet was in 1995 for user experience and honestly like genuinely one of the most core aspects to being able to make that in a way that a user never notices they're using blockchain is account abstraction except in our case we we baked it into the protocol into the normal M Pool so we actually have native account abstraction which allows even normal wallets like metamask you don't have to worry about what token you're making a payment in right if you send usdc you can pay the fee in usdc um but it makes every single wallet including EA wallets the smart wallet and that enables us a lot of things like

social recovery like no one should ever have to deal with seed Praises even people that have been in the space for a long time like store it in plain text at home you should never have to worry about that maybe your bank can be one of the signers to recover your account maybe your mom maybe your email account can be one of those and then if a certain threshold of those people say you lost your account you can recover it and there's ways to essentially stop anyone from hurting you in that situation as well you you can have transaction rules so for example let's say I can spend 50 a day without having to resign a transaction and this way you can do your normal groceries you can like play video games or buy assets without signing every single transaction

which is just horrible ux um we have already uh the secp of 256 R1 elliptic curve as a precompile what does that mean that means you can essentially just sign in with your phone like face ID or the the uh fingerprint scans or you can use things like Enterprise security like UB keys or web off to sign into your wallet it stores the seed in the secure Enclave in The Trusted execution element and you never have to worry about signing things or where your seed is is uh saved it's saved in a pass key and it's secure and you can just use your phone without having to worry about it right uh you can do session key signing

so let's say you're playing a web 3 video game in the future it'll just be called a video game by the way uh you can essentially sign into that game once and then you don't have to make every single move you don't have to sign another transaction because you have a session key and it lets you keep playing the game without signing in just once like you should um another like core feature of like this is not possible in optimistic L2 they're starting to advertise that you can have offchain data in like enan or Celestia but that's actually incredibly unsafe on optimistic rups because that da solution can collude and literally steal all of the funds from the L2 and it makes it immediately like

the biggest attack Vector those fraud proofs don't help you at all and the entire ethereum community and research Community as well as L2 be all agree that uh a optimistic role with offchain data is actually a side chain it's not a rollup at all it doesn't get any Security in this case from ethereum um and they actually like this the reason why they're doing this is because optimistic RS have to push all the transaction data for every single transaction on chain to get their security and what that means is as ethereum fees go up optimistic rollups will get more and more and more expensive like the same way that ethereum got to hundreds of dollars for one transaction that's going to happen

overtime to to optimistic grups as well so they know that they need offchain data because it's going to get too expensive and they they have inferior older technology so they just they're like pretending like this is safe and they're betraying their values with uh zero knowledge proofs you can essentially take the state root of the Merkel tree and you can push that on chain and then you can have a use a zero knowledge proof to um essentially attest to the state transition right in this way you can actually have data offchain it doesn't affect the security of the protocol even a little bit like or okay a tiny bit but basically nothing because even if that data colludes or someone

withholds it can never steal any of the funds off of ethereum and one of the cool things is because again the most expensive part of a rollup is pushing data onto L1 this allows us to essentially have flat transaction fees the as ethereum prices go up these transaction fees will be virtually the same it'll be fractions of a penny and we have ways to make it cheaper and cheaper where it's massively scalable uh we are Security First this is why we're an L2 this is why we build on ethereum this is why we use zero knowledge proofs and don't just trust some validator set or pretend like Economic Security can be safe um it's our first second and third priority

right just the last year alone we spent $8 million on audits in just era and that goes up really quickly we spend a lot of money on it because it's essential because all these like cryptography itself is a matter of time that how secure it is and a lot of people worry that for both optimistic grups as well as zero knowledge proofs that ZK for example is a Leading Edge like cryptography that's been only around since like the 70s um until it wasn't actually okay that's besides the point it wasn't non- interactive until recently but it's it's very new as far as mathematics and cryptography goes uh so people are worried that maybe we Implement our our Pro System wrong or

our circuits wrong or our verifier on ethereum wrong and that's a legitimate concern but what we can do is we can use trusted execution elements like Intel sgx so you can run the same exact State transition like all the transactions essentially through the Zer knowledge uh circuit and you can also run them on this just a normal chip that is used in normal everyday like Enterprise consumer applications including again the base ID is the trusted element um and this essentially allows if there's a Divergence like if there's something wrong with the ZK uh uh circuits essentially this will allow you to like Intel sgx in this case can just freeze the protocol temporarily and that'll

make sure that no user assets are can be uh stolen or lost and this is actually one of the things that makes ZK actually safer than optimistic and I know that's a uncommon opinion but essentially to hack this system you'd have to find a zero zero day vulnerability in our ZK proofs you'd have to hack Intel like you'd have to find a zero knowledge uh exploit in that you'd have to uh uh hack two of our security councils which are completely separate entities around the world like you'd have to hack governance on top of that and then maybe you could steal something before the governance could freeze the protocol like it's it's virtually impossible like even a state actor can't really do

this uh and again we were hyperfocused on being able to actually make blockchain scalable to the masses over five years so there's a reason why we are ethereum um um compatible and not necessarily 100% equivalent because we need this to scale that being said over the last year we've already built out support for all the major tools Foundry which is Amazing by the way remakes hard hat um and we are making a massive investment into user EXP exper developer experience in product because we're the first blockchain that can actually scale to real world applications so now it's all about how does your mom use it how does your friend use it how does no one

have to realize that they're actually using blockchain and part of this process for developers is what be coming out we have an internal project that we're investing in significantly that'll make sdks so every developer even if they're not a blockchain expert can make these beautiful applications with amazing user experience that can actually not just be as good as web 2 applications but better um and we pledge to go beyond evm like that native account obstructions a good example we did that years before anyone else and we are going to for certain things where it makes sense we're going to have to diverge a little bit from ethereum to enable these worldclass applications if ethereum doesn't move fast enough and we

will do this in an open source way so it'll keep contributing to back down to ethereum but we have to make certain steps to really enable this to get to a normal person and all of these like little things that I described State diffs validium ZK in general like like these were handcrafted by a worldclass team over five years to make it right from the beginning that's why you haven't heard quite as much from us in the beginning because we had to build the architecture to the point that we can scale we don't want any quick fixes we don't want to just not have proofs like for a while because that's irresponsible and dangerous and doesn't help anything

in this case these you have thousands of decisions every single day in engineering where like especially in a large larger organization or medium organization where it starts diverging more and more and more every day so I know it's hard to understand the difference of like zkl 2s one from another or l2s in general but these are massive differences that can't just be slapped on like optimistic growups know that they have to move to ZK but it they can't just slap on ZK they can but that'll absolutely limit their scalability to 2 to 30 30 TPS it's going to take a long time to make those proofs so any of this sharding architecture that I'm about to talk about it's just you can't do that no Bank wants to have to rely on this kind of horrible mess of

reorgs or or uh possibly losing all their funds with offchain data that isn't really a rollup you can't just slap on these decisions afterwards it takes careful engineering over a long time and so that's the core of every one of our hyps that's where we're at today but we've actually gone further than that again because even if you say that 10,000 TPS gets you somewhere it doesn't right so the the we we worked on and this was part of our vision from the beginning something called ZK stack and that's the framework where you can easily spin up more and more of these chains even if the core is different if the core is the same you can just quickly spin them up right but the key

part here for interoperability for making it feel like this one ecosystem this one chain one internet is zero LGE proofs there's no way to fake this like you can fake this but nobody in their right mind no bank is ever going to sign up for that risk no no like Enterprise is ever going to want to hedge that amount of capital loss or rely on every single other actor like again this removes the if you don't have zero knowledge proofs and those chains can't trustless interact with each other there's no point in using blockchain a bank already has to tr trust another bank why would they want to deploy on this instead they get limited benefits and it's not enough to get them out of bed zero knowledge proofs are essential it'll be the the sort of https of

blockchain and so Z or ZK stack which is my product um is a developer friendly modular stack that just makes it really easy for you to quickly boot up these um these different zeka chains that are all interoperable using zero knowledge proofs and as a Zain you have a lot of different choices that you can make while still being completely connected to this system so you can have your data and rollup mode or offchain data in Validia mode or you can have volition which is essentially both of them you have a savings account and a spendings account maybe it cost a little bit more um on rollup mode to make a transaction but that's your savings account and then you can have normal spending accounts on

validium where you can make lots of transactions for very little like you can choose different transaction sequencers um transaction sequencing layers to be centralized not centralized and again that doesn't actually affect the users Security even if they're using that and it has only one centralized sequencer uh you can have different data availability modes Celestia again safely again being the key Point not making it a a side chain you can have private data public data and you essentially get a lot of choices in between um in the future much much more choices with zki which is the part that I mentioned earlier for for having different virtual machines and stuff and ZK stack is what enables like

this is core this framework model is core to how you can become like limitlessly scalable again the internet can scale on one computer it scales by doing execution shorting Apple cannot figure like they can't fit their entire user database in one computer it's impossible it's it's just way too much processing and and this is how all of Technology scales is by doing sharding essentially sharding of execution sharding of resources and ZK and ZK stack is what enables blockchain to get to this without having those compromises that remove the whole point of using blockchain in the first place and ultimately don't even scale to the place that people need

so uh again we use Zer knowledge proofs to make these interoperable so One Bank does not have to trust another bank if you send if a bank says from China sends a bank from from Europe to Europe a trillion dollars nobody has to trust each other in this like they know for a fact that Bank of China had those Euros they're not lying about some digital bounds that who knows where it's stored uh they don't have to spend hundreds of Millions uh uh every single year on doing reconciliation on doing auditing each other on having these middle men that you can't have transactions over the weekend or overnight um that essentially steal money if you're doing

Forex Pavements from one country to another the the amount that they steal from you is outrageous but the reality banks have to deal with this too Enterprises lose a crazy amount of money in just doing these International settlements zero knowledge proofs and not having to trust each sh not having to trust each other is key to this each Shard not having to run the executions for every single one of those chains or those shards each computer is essential to being able to scale to that level and if you create this right if you create make the ZK proof fast enough which they're becoming you can have your funds on one Shard for example Kronos chain you can send your e to a completely different like uni swap chain

for example you can swap your eth to usdc and then the usdc comes back to you and all of this happens in one transaction without you having to need to know where is this Dex where is the what is what shart is my assets on like it just kind of works like you can abstract away all this in know uh front in uh interface where it just feels like one massively scalable Network and this is like again this is the point that creates this internet of value this the same way as Internet it becomes limitlessly scalable as long long as there's resources on the planet left but but this is like like the same way the internet works today you might have your money in some Bank you might have it on some app you can just walk

into a store you can spend it you can walk to ATM you can withdraw money you can like like you can use a uh uh some online store to buy something you can play a video game on some web server and none of this has to be trusted like you don't have to worry about how this interacts it just works and that's what Zer knowledge proves and the sort of EX ution shorting enables for blockchains as well without making compromises getting every piece of benefit from the blockchain and this concept is a little bit harder to to grasp but essentially again ZK is not our bottleneck you can make a proof of each transaction then you can take those uh different proofs

and make a proof of proofs uh to represent the block and then you can verify that in seconds then you can take all of those blocks make a proof of proofs again get a batch uh proof and then you take all those batches and it becomes one proof that Rec that represents that Z chain and then you can take all of those ZK chains make them into one proof and at some point ethereum will also be uh a Zer knowledge proof blockchain we're already working on it like the ethereum community and then all of a sudden you'll make a proof of every single ZK chain that is deployed on our shared Bridge um you can verify all of those on a mobile phone in Mise so you know for a fact that millions of transactions were executed

completely correctly in milliseconds like it enables the vision of like ultimate decentralization where every single phone can be a full light client your metamask is a full like client like this is a lot more exciting than than it's easy to explain and we call this concept Z Singularity verifiable computation ZK is the end game right and and and this is what enables blockchain to finally scale to get to the promise that we all wanted and the cool part of this sort of ecosystem that we have oops actually let me let me step back a little bit so we already have like this really exciting

ecosystem that's growing it's honestly it's a super exciting um time to be in in blockchain it's finally getting to the point of like similar to the internet in like 1995 you didn't have scalability before that and as more resources became available Network bandwidth Computing Hardware became faster all of a sudden you started having Google Maps like you had social media accounts that scalability was key to being able to enable these real product Market fits applications and we're already at that level like like I I literally spent this week talking to uh four different banks a central bank by the way they're all deploying on us um uh different Enterprise uses I mentioned cork ID

which is the Argentinian digital identity system uh in Argentina 30% of people use uh crypto for their everyday payments and purchases and the vast majority of them use lemon cash which is kind of like a PayPal or payment app um and they're building a a ZK chain we have the top web three social media which really enables a new paradigm in social media again harder that to explain than it sound but um it'll revolutionize the way that social media works and they had to choose ZK sync because it was the only option that could actually enable real Market scalability we have Kronos or crypto.com deploying on us with addressable user uh base of over 80 million people we have I

candy which is the largest game developer in all of Southeast Asia and Australia and New Zealand that are launching uh going almost fully into web three launching uh games like assets for their media franchise is we have Fidelity which is a massive like many okay I'll leave the Fidelity one for later you guys will see some announcements about that in the future but um so we have uh space and time which is verif like ZK SQL you can essentially have verifiable SQL quaries and Google's like really excited in them they invested in them heavily they're probably going to integrate it at some point into Google big query which is massive right they're launching a chain on us because they need a source of

Truth to have their verifiable execution we have so many of these projects that I like I I wish I could tell you it's like it's a really really exciting day today like we are at the like the preist like the Tipping Point before launching into Mass market adoption and there's very few times in anyone's lifetime if any in human history where you have this big of a uh uh like technology or innovation that's about to take over the world and and if you knew the day-to-day conversations that I have you would be just as excited like heartbeating excited of of what's about to happen um and again like important thing to realize here is like these are not

just some apps deploying on on ZK sync like like gravity is is a highly qualified team like these Banks like we announced the Deutsche Bank Singapore monetary auth Authority one they are incredibly talented uh teams and if let's say like ZK candy makes I candy their CH is called ZK candy makes really cool uh games then that doesn't just benefit them it actually benefits like crypto.com then they'll have more people coming on boarding adding money and the more people they add money the more I candy will go up or all these different um uh like ecosystem Partners it's it results in this sort of

exponential growth pattern it doesn't grow linearly to the amount of resour resources or or like in this case I mean like Financial assets or users it's it creates this sort of exponential growth pattern that takes over the same way that if you've ever heard of network effects the internet grew with network effects which is an exponential pattern everything we make at matter labs and this is sort of also a constitution for for ZK sync is is like we are ethereum aligned because we believe that blockchains need to be it's not even believe it's like the whole point of blockchain that they need to be secure decentralized everything we will ever make will be completely open source free to use Fork it copy it build on it like

we're we're not gating it it's for the public good every part of it is permissionless again you can forget any of these ZK chains at any point when they want to divorce the protocol can simply Point their like contract and move the funds to another shared Bridge like they don't get trapped in this ecosystem no one has a can stop you as a user from withdrawing your funds from one of these ZK chains like it's it'll all be permissionless and like please join the future like you guys are the next generation of Builders like this is like this is a really rare time where you can get into an industry you can be like very casually hanging out with the like essentially the equivalent of app

like Steve Jobs in a conference like it's it's you don't have to take yourself too seriously you can can like you can dress however you want you can make silly jokes like you can have memes and presentations and I come from the data center industry that was not the same um like please come build on ZK stack or just ZK sync as a Dap like this this ecosystem is about to grow very quickly because we built the fire and we're ready to essentially ignite that fire to grow and all of our ecosystem partners are launching so it's a really good time to come in for those of you that use uh I still want to call it Twitter X Twitter X um you can follow me at ZK Lumi um and for any sort of

official announcements on ZK sync you can follow ZK sync at ZK sync um if nothing else like the thing I want you to start thinking about to start taking away is like zon knowledge proofs is a new type of cryptography it's a new type of mathematics it goes Way Beyond uh blockchain like like anytime we we came up or found a new field of mathematics it changed fundamentally how we understood the worlds behind us it fundamentally in like change what we can build maybe it'll start actually explaining parts of quantum physics or black holes like this is it's it's math right like you should be incredibly excited this ZK is going

to be like same way as the little lock bar on your browser it's going to be embedded fundamentally into the internet into your normal identity Solutions it goes so far beyond blockchain it's it's math like like when were you last around when they not really invented a new field but essentially invented a new field of mathematics um and again with the permissionless or open source uh feel free to download it forget share this presentation do whatever you want to it um that's it and we can take some questions as well now if you guys have any questions on this uh thanks uh Daniel for this lecture it it's really exciting to see all the innovations that are happening

in ZK a question I have is like can you elaborate more on how exactly ZK is uh helping with the bridging between these like separate shards like how is interoperability uh powered by ZK yeah so essentially you can like if you're doing cross team bridging these are non-deterministic programs and by the way the the the cross chain bridging is the same thing optimism will do they just use fancy terms for it essentially blockchain a can never know the state of blockchain B because they're non-deterministic programs there is no way for them to know what the state of the other chain is with ZK you can essentially run that computation and then cryptographically mathematically

prove that that state is completely correct right but even then like there's a lot of now going around like ZK cross chain bridging it's like okay you can have a zero not proof of the state transition position but what's the correct State like ethereum might just Fork like maybe someone can literally Fork ethereum on their own computer and say double spend over and over again then be like make a real like verifiable zeron not proof showing I have all the eth and then you could steal the money from fridges if if you're doing source of truth that's cross chain you essentially need a multisig like you need a small validator set that ATT tested the correct state which immediately removes the point of like blockchains right because you have a

tiny validator so so what we do instead is we have one shared contract on ethereum and every single like that contract holds all the uh funds so when I send uh tokens from era to ZK chain C I'm not actually bridging anything it's like accounting the same way that any blockchain works if I send funds to somewhere else you just Mark that it's somewhere else that's how the ledger works right that's how our um interoperability works too and the way that you need to do this is like again chain a can't cost chain B so essentially you commit off of chain B to the to the batch to the new EST State on ethereum L1 which becomes your ultimate

source of truth right and then chain uh uh C can read that state commitment and verify the zero knowledge proof that this is really the correct State and you know the source of truth because the source of Truth is the same if ethereum reorgs every everything reorgs on top of it if one L2 after the zeron proof can't reorg and hurt the other like chain this is Zer knowledge proof that verifiable execution is essential to being able to do this trusted execution and it's also trustless execution and you also need a source of truth it's critical it's essential and you can do source of Truth bridging like if we do this across like

cking to to polygon all of a sudden latency is going to be incredibly slow it's very expensive transactions to do cross chain like you need to that's that's bridging in this case it's accounting right it costs you essentially next to nothing to send funds from one one chain to another and you can make them really really fast in the beginning somewhere around a minute and as zero knowledge proofs get uh faster and faster which they are getting faster really quickly essentially it'll become almost instant all right that makes sense instead of waiting seven days for every transaction to confirm which optimism will fake but again this is not going to this is not going to scale um the the

money required for it the risk that everyone has uh both financially relying on every single person and it's the weakest link problem again or or literally just like like not like if you have interoperability in this optimism system you're back to the problem of one computer needs to reexecute everything every single person has to have every single node running to have any source of Truth and even then it's like you're relying on that small validator set to not steal your money or lie especially if the data is offchain do not do offchain data don't do cross chain bridging yep that makes perfect sense uh yeah I don't think I have any other

questions yeah I wanted to ask one question uh yeah you mentioned that we can run any VM on these hyper chains so you would be able to not right now yeah okay so uh like if I have a VM that has like like different op codes like whatever like arbitrum is doing kind of they are using was based op codes so would we be able to run that kind of VMS because like that will change a lot of how the like VMS would work like normally the VMS would only have changes in their application on their or their in time architecture but since now the

op codes are being changed the it makes a lot of difference so first of all IP trims like I like The Stylist idea like it's it's really cool it's not quite as good as they advertise though they literally have live demos where it's like they try to copy a cryptographic library and it just doesn't work right so there's more intricacy to it than it seems but essentially so what you only really need to across VM you only really need to protect certain like certain operations like if if a person sends someone money you need to protect that operation because otherwise they can send it over and over again double spend and steal money from a different BM right so that we can essentially make a

very minimal set of requirements of like you can't double spend you can't send the same funds twice and then essentially you can use any virtual machine on top of that only protecting those certain things so and then you can do this verifiably and and this way you can have Solana VM completely trustless on this network without anyone being able to steal anyone's funds you can have move VM it's it's more exciting than just like DM right like I could have my own account as my own rollup but in this case now I'm only running transactions for myself right so I could have a literally a rollup that only defin the count um uh like logic it doesn't have to have every op code every

pre-compile it doesn't have to have any of this it just needs to update like key Pairs and do certain operations like sending and you could have this tiny minimal State minimal functionality roll up that by the way now is much faster to ZK prove and then I can interact with any other application on on the network like it's not as small as VM this is like when when we talk about ZK as the end game this is a large part of what we're talking about and this is not something we talk about publicly yet because we want time to essentially like really flush this out the ZK IP thing is very far like we know how we're going to do that we know how we're going to do all of this we just need time for engineering like this is not just going to create the next generation of

scalable blockchain it's going to create the next generation of like like blockchain as a whole all of a sudden like you can use an internet you can use net or you can use WordPress or you can use whatever you whatever you want this is what this enables for blockchains it doesn't matter what virtual machine you have it doesn't matter what language you're writing in it doesn't matter what type of rollup you have it becomes this massively let me get that TR back up it's it's a good way of visualizing it every one of these little pieces can be completely different but they can all talk together and and interoperate so it really creates a whole new paradigm in blockchain yeah sense

than no problem and again you'll be able to write in rust or C or like whatever language you want the virtual machine can be written in just about any language hey hey Daniel I had a question basically uh so you mentioned about recurring payments uh using the Z St so basically currently how it works is uh like uh in centralized Banks basically they collect the companies collect our credit card data in order to process the recurring payments so how exactly it's going to be different uh using the ZK stack and like we are not going to uh like lose our uh like Financial security credits Etc so this

gets this is a loaded topic there's so many different use cases in which you could do it but let's say that like let's say you took a loan for I don't know a nft a car or or a house essentially you can pre-sign and you can have conditional logic where essentially once a month only once a month this entity can trigger a move of money from your account and it can only be this amount or less and they don't need interaction from you to do that right and the the how State diffs makes that cheaper is a complicated topic that let's not go into right now but that's that's part of the the native account abstraction part as

well so like this is just one example there's many different ways that you can you can design this right you can have like if if you like lose access to your account maybe you can withdraw like one tenth of it a month and then it would take someone 10 months even if they somehow corrupt Your Guardian set like they corrupt your bank and your mom and your friend or whoever right it would still take them x amount of time like like 10 months to be able to take out 100% and at that point you could have said essentially remove them from the validator set like that's a sort of reoccurring payment too there's so many different Avenues you can take the recurring payments to in a way that's trustless and doesn't just screw over the

user yep that's great and we do have five more minutes if anyone has any any other questions okay well that's it um yeah it was it was genuinely a pleasure talking to you guys I can't wait to see the things that that you'll build um hopefully on CK sync I'm not going to push that like like blockchain as a whole if blockchain wins I'm happy I don't care who wins we just happen to be a year or two minimum ahead of every other player so by the time they catch up on trying to implement zero not proofs which doesn't happen in a year by the way um or or try to get that 3 TPS to anything actually usable like will be

much ahead of that so it's it's a good interesting place to build but I really just hope that we can build a better system where like the financial like the world economy is safer doesn't rely on one country lying and possibly breaking it like we're trying to get that freedom prosperity and and like we need people like like you guys especially since you're young to think about it differently and build a new world you're going to be the ones in power in the future you're going to be the politicians or or just Builders the next next like massive Enterprises you can change the world and you don't have to change everything you just have to change your own habits and then if enough people make small decisions it becomes this wave that's

unstopable yeah Daniel thank you so much for having this lecture it was a pleasure hosting you and hearing all these great innovations that are happening on zikas sing and in the space of blockchain amazing thanks so much guys talk to you later