Compliant Privacy in your wallet

Transcript

here guys. [applause] >> So, thank you so much to be here. >> No, thank you. Thank you so much for organizing. I don't think we ever had East Bilissi before. So, Ali was a great volunteer for this effort to happen especially with Ethereum becoming the home for privacy and I think that's the right time to do something like this. Um I will do I mean now we see lots of interest in privacy and usually when we see such interest it cascades to lots of events such as you know adoption of the vector. So I think the uh sector had lots of problems right it started with compliance problems before then with

usability problems because it was really hard and clunky to use privacy and the need for privacy was always growing such as you know like the user experience right so now we are in this overlap when these two lines basically intersect and finally forming the need and making private privacy reality. Of course, like there's always, you know, some government caveats, right? So, how to make sure privacy is compliant, but I think we finally reached the right place and home uh for your assets on chain.

Um, all right. So, HKO is a private home for your assets, right? And here I want to uh break down privacy in two different verticals, right? First is uh finance, right? like one big shift we see that finance is moving on chain right it's a huge industry uh lots of chains are focused on finance like Solanas uh you know and value proposition is inter internet capital markets right what I like in Ethereum is uh that it actually thinks more than finance it thinks about you know like attention uh like how to prevent you know governments from intruding your privacy Y and you know it just more global

vision than just capital markets right. So uh we have this like privacy actually has these two forms which is first is privacy prevents you from attention economy right so if privacy works well in the future it will finally kill attention economy so no one can sell your data you know Google Facebook and other big platforms won't be able to own your data so you won't become victim and you finally will control your attention at scale Um and the the other one is uh your digital finance your assets. So today we'll focus on this and hopefully in five years we will speak about cyber

punk and this attention economy uh killer as privacy on chain. Um so what we see now is that traditional finance is becoming obsolete right. So the biggest companies such as Stripe and Alibaba are launching their own chains and what it implies is that like Visa, Mastercard networks, you know, banks globally are becoming obsolete with their systems. We have lots of branches in banks, lots of people working there. So I don't expect that we uh they will actually adjust to u like crypto generally. Uh so my expectation is that fintech will

actually kill lots of banks if they're allowed to do it uh with um like the with the governments, right? But this is unavoidable because um banks won't be able to adjust. So ultimately new entities will be created and these entities will have licenses and these licenses will enable them to create crypto banks. Right? And we see this with Arabore in the US now uh with Palmer Palmer Lucky and Peter Thiel and you know lots of other efforts such as stripe right I'm sure that stripe tether will also acquire banking licenses and they will become banks right and um in

this world when uh banks move to move to chains and new entities are created to completely replace traditional financial rails. One of the biggest problems uh that still is happening there is privacy, right? So whenever you move your assets on chain, you can immediately see how much you have, how much your friend has and everyone else, right? Especially when it h when it's happening with peer-to-peer transactions. When you know your counterparty, this creates one of the biggest problems because now it's not just pseudonmous addresses, it just your friend and you know the history of

transactions and and what what they have as well. Um in the US we have two regulatory clarity uh now acts is genius act and stable coin act. Stable coin act was already passed and now it's about uh genius act uh to get full permission uh to basically enable everyone to launch stable coins uh and to move assets on chain. So with this bottleneck there are three ways you can solve for privacy right. So first is um you know which is quite popular now is a chain level privacy.

The problem with chain level privacy is that uh you sacrifice scalability uh and you sacrifice lots of UX primitives such as public AMM pools and others right so if you so basically you end up launching very slow chain uh and that that's a big sacrifice in my opinion uh plus you don't have any network effects right so to bootstrap layer one chain it's incredibly hard right who was able to do it probably three chains ever. The race chains, the rest of the chains are called ghost chains. Uh so no one wants to become ghost chain. Second approach to privacy that was quite popular before with

tornado was up layer privacy and it implies that you sacrifice capital efficiency of your liquidity and user experience. So your assets are stale onchain. So they are not active. you cannot do anything and then you have to do lots of clicks to access this user experience and we had this before. So our first product was actually uh the dub and we had 250 million volumes for the dub but then we realized that this is not scalable approach. So the most scalable approach for privacy is wallet level privacy on public chains. So with all the advancements of uh technologies such as zk addresses tees

as well as f uh it it became possible now that you can actually almost don't feel privacy while accessing any type of transaction on public chain such as ethereum and sana. So hlel is a gateway for full fully self-s sovereign and private money. You can move onchain directly to HKO private address and your assets are sitting in the shielded pool with UTX source and you have your own private address inside HKO. It implies that you can make any transaction in full privacy. You can pay your friends in full privacy. You can do yield farming or trading in a full privacy.

Proof generation um is client side proving so it's fully secure but you have the choice to do um proving in TE right and this way on weaker machines such as your phones you'll be able to generate zero knowledge proof and make the transaction instant right that's how we solve the problem of um instant transactions right that usually uh zk has what appears onchain for all transactions is HKLE smart contract. Um and and basically like you can have sender recipient both um having hle addresses and what appear for this type

of transactions is sender HL and then recipient is also hle smart contract. So this way we avoid full like any traceability of your money onchain and you know think about also like current big trend which is cryptocards. So cryptocards are you know are going to actually take take the world because you they you can earn way more money with crypto accounts. I have 10% APY on my current account, 3% cash back with Etherfy card. But the problem is whenever you send money, you have to KYC there and and then you basically show your face and link your ID with the history of your transactions, right?

That's one of the biggest problems that Hinkles also solves, right? So we're partnering with Etherfy there. So you can send any asset you know USDC ETH uh to Ether without revealing the history of your transactions and spend your money without any traceability uh with a proper hygiene. So the product suit that we have is first the most one the the most important one is HKO wallet right it's a fully consumer wallet it applies to traders applies to people who want to own money privately who have you know shitty governments as well and don't want governments you know to have any exposure on their activities right and

uh um and then just basic consumers who want to pay in crypto Uh second uh product that we are developing right now is HKL multisync and this mostly for institutions organizations. So you can connect with your multisk or MPC accounts such as fireblocks and others and then you basically activate hinko transactions with their signatures. This allows us to access uh kind of like bigger audience in terms of u TVO because you know most of money in crypto is it's actually in the hands of few uh bigger whales and the last one is hinko for developers

right so the problem we actually started with developer suit so APIs for payment networks but the problem is that if people don't understand privacy or zk right they won't be integrating your API and SDK. That's actually what triggered our decision to build Hinkle Wallet. So with Hinkle Wallet, we can actually show everyone how easy it is to access privacy and how huge is the problem. So this is our next step where you know we actually now we released Hinko wallet. It's fully public everyone can use. So now we can actually introduce SDK for developers. So any payment network or payment solution can integrate privacy uh within their

system. Uh now we are getting to the you know heavy brain part. So prepare your [laughter] um like my co-founder brother and CTO Nikico will introduce. >> Hi everyone. Uh we will go through technical details of our protocol. Uh my goal here is to explain you how we built Hinkal. Uh at least give you some recipe how to build it. As Andre Karpati says, you do not understand something unless you build it. And this is my approach for you to understand something in our presentation from technical perspective.

So uh when we build hallelu essentially we build information system which is based on some primitives. uh in our case it's cryptographic primitives because crypto cryptography is used for privacy and anonymity to achieve uh to to achieve compliance even we use uh cryptography. So I will explain some basic u functions by some basic primitives that we use in our protocol. If you think about Ethereum, Ethereum signing, Ethereum encryption or hashing, Ethereum uses some different kind of primitives like maybe if you think about Kchak or SH 256, it's a common hashing algorithm on Ethereum and uh ECDSA

signature for signing transactions on Ethereum. So this is very common standard uh cryptographic primitives used not only on Ethereum but generally everywhere in the industry. But when it comes to ZK, it's not the same essentially because uh ZK uses little bit different math in addition to normal computation and it doesn't like some kind of u operations. For example, it doesn't like if if conditions somehow it doesn't like and uh common primitives that is used generally in the industry such as sketch uh are not applicable for ZK because they are computationally heavy because you need to prove something in ZK to the chain and chain

should verify you. So that's the reason we picked uh some non-standard uh primitives and if we're talking about elliptic curves it's baby jab job curve uh it's baby because it's little bit reduced security of the system but allows you to prove efficiently statements in the zk uh here I will I give a description of the curve I will not go to every detail but I will cover the high part second one is pad and hash function allowing you to prove in ZK some hashing statements for example that uh offchain UTXOS have generate this commitment you prove in ZK by showing

that this uh this UTXO gives you the specific hash uh and for uh generally for ZK we using GS 16 gross 16 is the golden startant in ZK it's one protocol that was introduced in 2016 by Jens Cross and still even now it's not written in all parameters. So it's kind of cool standard that many protocols use up to now. Uh we are using it uh and for growth 16 it's implemented in circum. So circum it's a library that allows you to generate provers where you prove offchain some statements and also circum allows to generate you verifiers.

These are smart contracts that you deploy on chain and you verify statements. I will talk about what kind of statements you verify it in the next slides. First of all, uh if you think about uh privacy, you generally cannot consider only blockchain as information system. You need to move step up. You need to consider users who are using this blockchain because we are talking about privacy of transactions of these users right so they are off the system of the blockchain. So we should consider little bit bigger system in this case and uh the key I mean the recipe here is that you should have some kind of

information some piece of information that is only known by you and it should be stored offchain. In our case, it is stored in our wallet and it's a shielded key. We call it shielded key. It's additional set of keys that is stored offchain. Nobody sees except you. And with this shielded key, you allowed to do shielded transactions that I will explain later on. Uh just one second. Uh this shielded keys uh split essentially into viewing keys. No, go back. viewing keys and uh spending keys.

Spending keys allow you to spend funds that belongs to you, shielded funds that belongs to you and viewing key allow you to essentially view all the funds that you have. You may ask why we separate it. In Ethereum, it's not separated. It's the same. But Ethereum is a public network. Here when we think about privacy we should be able to differentiate ability to spend something privately and ability to show others these funds. For example, you want to prove the government that you don't own more than $10,000 in shielded assets.

You can do it with just giving viewing key. So when you give viewing key you don't give access to the funds but essentially you give access to view the assets. So that's important distinctions that allows you to differentiate these two functions. Uh let's talk about stealth addresses. Uh stealth addresses is not a completely new primitive. Uh essentially they originate from defy helman uh asymmetric uh cryptography. It's a generalization of the cryptography. I will remind you for defy helman we have one private key and one corresponding public key.

Uh and uh stealth addresses is generation of defy helman algorithm protocol when for one private key you have multiple public key. So uh if you look on the left this is a canonical public key it's just one point on elliptic curve but for stealth addresses you can generate h0 and h1 this is second public key so you can generate any number of this public key any finite number within the field and this allows you to achieve uh privacy when the outcome of the transaction is uncertain. Think about uh when you want to swap one token for the another token,

you never know what will be the outcome exactly because there may be some blocks before you. Sorry, the transaction within the block before you that may change the prices and your outcome may be different. Stealth addresses allow you to capture this uncertainty in the outcome such that users who submit this transaction remains private. very important concept here. Uh who is familiar with Zcash the protocol? Okay. Okay. Good. So, Zcash was one of the original uh pioneer in privacy and it introduced this system called UTXO

system is unspent uh uh transactions and uh this term essentially comes from the bitcoin like every node every amount is UTXO and when you spend it it's nullified. So this is something familiar to you this UXO system and it comes out that this UXO system is really important for privacy. It allows you to essentially store your funds and then claim the ownership of these funds later in ZK without revealing who you really are.

Think about UTXOS nodes and on in on each nodes there is a owner of this node. For example, five USDC node belonging to Georgie, but nobody knows that it belongs to Georgie. What Georgie can do? He can prove in ZK that he has a private key that generates public key which is inside this note. Once he proves this statement to the smart contract, smart contract releases this 5 USDC and tells Georgie, you can spend this 5 USDC however you want essentially.

When this 5 USDC is released, you can essentially call any protocol unis swap in our case and swap this 5 USDC to 5 USDT. But when you want to deposit back this 5 USD to the shielded pool. So you don't link your identity here when stealth addresses come to the play, you provide stealth addresses. So this is a one-time address which doesn't reveal your identity. The beauty of stealth addresses is that you can generate billions of stealth addresses and all of them will be unlinkable for the outsider. So the same person can do any type of transaction and all these transaction for the outsider will be unlinkable with each other. That's a

beauty of this uh def helman extension. And the last part this is a new that's not new part but somehow we should account for all the balances in the system utxos and the common way is to do it with miracle trees. So essentially we should know the state of the system. It's a balances of all people in a smart contract and uh what is essentially actions they are doing and we need to capture it with just one number and uh this number if you think about cryptography is generally hash hash of some data data can be arbitrarily long

but you still can hash it and prove integrity of this data later. uh you you may ask me like why don't you use generally one hash function and essentially hash all the data but the problem is that when you you when you do transactions you remove and add UTX source to the system and you need to do it efficiently because if you rehash everything you need whole previous data which is really inefficient. What you need to do is to do it in a very short time and if uh you have n data n number of data the shortest ways to do it is in log n operations and miracle tree allow you to

do this. So by insertion if miracle consists of let's say 16 you can do in four operation. So quite fast efficient way to prove integrity of your data and to record integrity of your data. Uh we have two types of miracle trees. Uh I want to stress it out. So if we think about miracle trees uh the main questions that we have do we need to hold the whole Miracle tree or we can hold just one part of miracle tree and the answer to this question depends what do you want to achieve. If you want only appendable miracle tree when you only can add some leaves then the answer is

that you can essentially store log n elements is the most right leaves for every level and then update this components for every insertion. Insertion just adds another additional leaf to the miracle tree. And the second point is about uh the case when you want to remove miracle tree uh component. Think about the user who essentially deposited some asset in the hingle but later somehow we find out that this user is a hacker stole the funds and wants essentially send this funds to centralized exchange. In this case system should be able to heal itself and

to block the access of this hacker to the funds that he stole. How can we do this? We can do this with removable Merkel tree. So we should remove the access token the leaf of the hacker from the Merkel tree. But to do this just the right leaves right elements are not enough. You need to have the whole Merkel tree. And by removing just one component you need to update the upper hash of Merkel tree. So that's a distinction between uh incremental Merkel tree and full Merkel tree. So in case of access tokens for access gated hingle we use full merkel tree and for commitments and utx we just keep incremental miracle tree

>> things to remember for nontechnical people based on what nika said is baby jab jab zk addresses at ber trees um and now we can move to compliance and security so we started actually hinkle right after the cash sanctions and one of the biggest problems we had is to persuade um law firms and to to to make a you know legal memorandum that we actually have can have compliance with privacy right it was quite tough to do but you know ultimately the system that we built around privacy is fully compliant with governments so we never try to avoid any governments or do

something against them um and at the same time it protects people, right? So we don't have any special keys that allows to reveal transactions, right? So it's a full privacy of your data. And the way we achieve this is twofolded. So first is when people interact with um the wallet uh we actually check if the wallet you know that they're depositing from is um compliant like and for this we run uh the compliance check with chain analysis right so if it is red flag wallet um it's actually not allowed to interact with smart contract so the

transaction is reverted right so that's a first kind of checkbox u to make sure that we are not allowing illicit wallets to interact with our core smart contract and you know core privacy primitive and then if you know it's and this is an entry point right what happens after the entry so for example you uh you were not red flagged suddenly you became red flagged right and you're already inside Hinko and for this type of use case uh we actually allow users to revert transactions So basically to lose privacy uh and to send back only to the main address uh he deposited this asset from right. So this allows us to prevent

uh the user from actually using privacy primitives with us and at the same time you know making sure that he he is getting the asset back if uh he you know if he has nothing to do against the law right so lots of actually chain analytics tools miss and you know just misinterpret the data and so they you know they can red flag you because uh like the centralized exchange that you're depositing from um got some contamination from some illicit other illicit users, right? So, you should not be uh reliable for and in charge for someone else mistakes, right? Um and the

second primitive that we run for large deposits is uh um like this kind of primitive that is called um reusable attestations, right? So reusable stationations is um basically allowing you to prove that you did KYC somewhere before you know sometime before somewhere in internet such as you know it can be bank of Georgia in it can be you know revolute for Europe or Venmo account for the US right so what you do is you generate zero knowledge proof of ownership of your account and then we accept this proof but you don't share any data with us. So it's a fully

private way to prove that you actually have information. You actually did KYC sometime before without revealing uh this to us. Um in addition to this we have you know bunch of other cell bone tokens from you know Binance, Galaxy, ZK. If you did KYC with them you are immediately whitelisted. So you can access um the protocol with any deposit. So this rule triggers actually for deposits over 1k of each asset. So if you have 10k of different assets, you don't have to KYC or provide uh any reusable attestation to us. Yeah, we have a team of uh 18

people actually now. Um you know my background I start as professional wrestler. I was wrestling Sambbo for 10 years. I was twice a world wise champion actually in Sambbo. Then I uh spent five years investment banking. Uh have went to Stanford did two degrees MBA and masters there and during um my second degree actually we started HKO. I'm al also the founder of Silicon Valley icons. It's the biggest Stanford founder community. Uh Nika has CTO his PhD from University of Warwick and also built few apps. We we built few apps before actually both in blockchains and web two.

Uh we always hiring so we have everything uh for great people to join. So we are entry point to blockchains right to the new home for finance. Um there is also the vision beyond finance where Hinkle is also entry point right where that kills attention economy and allows public blockchains right to be the new ledgers for this economy with the ownership of you know your assets. We have all the great technologies. So on boarding is a bit longer for non-technical people because it's pretty complicated tech right and you have to understand it but it's possible. Um and then we pay great salaries every six months promotion and bonuses.

Uh we raised over 6 million in SID round. We are planning to raise series A soon. And yes, own your money with HKO ho.pro. Here's my telegram and email. Feel free to reach out if you have any questions.