CTRL/SHIFT - 14 06 2026 PARTHENOPE STAGE - MIKE MCCABE - ONCHAIN PRIVACY IN THE FUTURE

Welcome to my talk about the future of on-chain privacy. Um Let me just dive in right away by saying that like privacy is a very wide stack and it exists in many, many different layers uh within even just a blockchain space. Um I do not know a lot about all of these layers. There's basically one layer that I feel qualified to talk about and that is the application layer and the compliance and identity layer, but I can quickly go about the other ones as well. Um so we have like the network layer, right?

And that is just like how you configure your computer to access the internet really. So like if you are not using a VPN or whatever, like you are already leaking a lot of privacy. Um then you have like the RPC and data access layer and that is then how you from the internet access the blockchain. Um again, like here you know, maybe know when you're in MetaMask, you can like configure your RPC and these kind of things and like a lot of RPCs do a lot of different things. Um there are actually quite a few good overviews on like what RPCs preserve your privacy and which ones don't.

Um it's definitely worth looking into, but I think Infura, the MetaMask default, is already pretty good. So we're actually not doing too bad on that front. Um then we have transaction ordering and the mempool layer. So a lot of you might be familiar with like protocols like CowSwap. Um CowSwap is basically like an encrypted mempool.

What a mempool is is when you submit a transaction to a public blockchain like Ethereum, it is queued in like a waiting line, you could call it. That waiting line is called the mempool. Everyone can view this mempool and so what happens is you have like MEV, which is like I think maximum extractable value or miner extractable value. There's like a huge discussion around what it stands for. Um these kind of like searchers, these bots, can kind of like see, okay, with these are the transactions coming in, uh I can order or I can do some manipulation on these transactions to make a profit for myself.

And so, that is also like if you don't have a private mempool, you are also leaking privacy and you are getting you're paying the price for it literally. Okay, and then we come into the application layer, right? And that is where I have privacy pools work in and spend most of my time at. Maybe wait. Yeah, maybe start with this one.

So, if you look at blockchains and where we are now, we are currently in the stage of pseudonymity. And that essentially means like when you sign up for a wallet, you get like a random address which is like 0x whatever. And that basically is is a pseudonym for your identity. But what happens with that address is that everything you do on it is public and will remain public forever. And so, if you like combine that with a lot of like data breaches and like centralized exchanges and general privacy leaks and like the possibility to just like identify information from your transaction history, it is really quite simple to identify who is the person behind a specific address.

You know, like Coinbase, Binance, they have all had specific leaks. It is again, yeah, not not very hard to find the person behind a Coinbase funded address. Then you have confidentiality as well. And so, that is like a specific token standard where like instead of using an ERC-20, you can use a specific token standard that allows you to for instance hide the amount you send or hide the recipient that you send to. But it is like partial privacy and it is still happening on chain.

So, where we want to go to with privacy pools is actually anonymity. And so, you don't know the asset, you don't know the sender, you don't know the amount, you don't know the recipient. The only thing you know is that something happened because there is an on-chain record that something happened, but you have no idea what or how. And so, the way we do that is through zero-knowledge proofs. We use Groth16 SNARKs.

Um and then the other end of the spectrum is like fully private um fully private chains such as well, I think Aztec is the is the biggest or the best example of this because they have private compute along with like Aleo and Taiko. Uh but they are not compatible with Ethereum or public chains as a whole. Um and we still want to focus on Ethereum. So, uh that's why we're in the anonymity sector. Okay, let me go back a bit.

Um yeah, so where are we going with privacy on Ethereum, right? Uh there's a lot of talk about it. Privacy is now like part of this crops mandate. So, censorship resistance, open-source software, privacy and security. It has really become like a first-class citizen on Ethereum, which is really exciting to see because basically like since Tornado Cash happened, everything that will have anything to do with privacy was basically considered for criminals.

Um which made building in the space incredibly hard and which has also set development in that space back like numerous years. Um so, how is this going to happen? Well, the EF has already put in their roadmap that they want privacy to be native in the uh execution layer. So, how that is going to look like we don't exactly know yet. It is could either be ZK wormholes.

So, that is like a a a an EIP that has been submitted a few months ago, which is like really interesting. You can essentially just send ETH to a burn address, which looks just like a regular address, but we know or the protocol knows it's a burn address. And then you can mint ETH to a fresh address. So, this is like a really cool way to like possibly deniably use a privacy protocol. Uh then we can just have like shielded pools, which is what privacy pools and Railgun for instance do.

Um or yeah, there could be whatever. I mean, the key thing to know about this is that I think that privacy natively on the L1 is going to be at least three, four, five years out from now. Uh the key reason for that is is that introducing it natively in the in the L1 right now would bring along a lot of risks because ZK proofs especially on chain have not been around and tested and battle hardened to the extent where you could probably say like, "Okay, we can now implement this completely risk-free." You only have to look at the Orca pool exploit from 2 weeks ago or so to to look at that in practice. Um so then I'll look at the current state of privacy on Ethereum.

And spoiler, it is absolutely dire. Um I say like 0.01% of Ethereum activity is private, but if you actually like look at the numbers, it is something like 500 transactions uh of the 3 million transactions on Ethereum a day are private. Um so that is like 0.00000 something.

Um so it is really like a space that has a lot of room for growth and a lot of room for improvement, and that is exactly what we're doing with privacy pools. Um So, this is so interesting. So, yeah. What we're doing with privacy pools is essentially um three things. So, the reason why we think privacy hasn't taken off is these three issues that we see here.

First one is as I mentioned before, the compliance risk. With Tornado Cash, Samurai Wallet, we've seen so much hostile regulatory enforcement against privacy protocols and privacy dev specifically that I feel like this in itself has stalled development of privacy for like many years, right? Like you could even see like Aztec for instance launched and then Tornado got sanctioned and then Aztec de-launched basically. They shut down their infrastructure. It has been like a real guillotine almost hanging out of hanging over this this particular space of the industry.

But honestly, that's not all because I don't know how many of you have ever used a privacy protocol, um but it is incredibly complicated. You need to have a very high understanding of blockchains. There is a high knowledge bar to use privacy protocols. And so currently, it is really like exclusive to the cypherpunks that are willing to invest their time and and effort into learning that and that is not really something that you can expect of users if as we want it wanted to be like privacy is normal, it's not going to happen if if it is so hard to use. And then thirdly, we also have like the UX gap where um privacy protocols as they exist today are incredibly limited in their functionality.

Like you can basically like deposit and withdraw and that is all. Um but if you look at like what people want to do with their Ethereum addresses, they want to do DeFi, they want to do lending, they want to do all these different possibilities and that is just simply completely incompatible currently. Um so yeah, I think these are the main drivers why privacy hasn't taken off and why privacy is not normal on Ethereum currently. And that is exactly what we're trying to tackle with Privacy Pools V2, which is what I'm excited to tell you guys about today. Um so firstly, the compliance risk, right?

Um so this is actually uh largely addressed in Privacy Pools V1 uh with our ASP, which is like the association set provider. I'm going to tell you guys a little bit more about that in a bit, but basically the history behind that was um uh Amean, uh one of our co-founders, uh was the first funder of Tornado Cash. Um and then when they got Tornado Cash got sanctioned, the devs got arrested, he sat down together with uh Vitalik and Jacob Illum who is the chief scientist at Chainalysis, and they published this white paper um for privacy pools, which is like the uh association set provider uh paper, uh where we essentially introduce this new concept, which is the proof of association. So, essentially what Privacy Pools does is when a deposit comes in, we check the source of funds and we trace it back to its origin. And so, if we see that that is like a legitimate source, like a Coinbase, a KuCoin, a Binance, whatever, um we will accept it into the pool because we know that these kind of exchanges have their KYC practices and AML practices in order.

Um if it comes from an obvious illicit source, like scams, exploits, whatever, um we have the ability to like deny them privacy within the pool. So, we remain non-custodial, we never take any custody of their funds, but the way it works is that in order to gain privacy, you need to use the relayer within our protocol. So, our relayer essentially execute transactions on behalf of you, and that's where the anonymity comes from because the relayer submits it instead of you. And to use the relayer, you need to have a proof of association, that essentially means a proof that you're on the allow list. If you don't have that, you can only withdraw back to your own depositing address, and so you don't gain any privacy from the protocol.

Um then we have the integration problem. And so, what our goal is for our Privacy Pools V2 is essentially being able to deprecate our own front end entirely. We have free SDKs that come along with Privacy Pools V2 for different kind of use cases. Obviously, we have the Cohaku SDK from the Ethereum Foundation, which is like specified for wallet specifically. Um that includes Railgun, uh, privacy pools, and Tornado Cash.

Um, what is cool about the Kohaku SDK is that it doesn't just look at the application layer, it looks at like the RPC layer, uh, the mental layer, all these other stuff as well, and so it's like a very comprehensive privacy stack. Then we also have a TypeScript SDK, we have a Rust SDK. Uh, the goal of that is essentially to have embedded traffic through flow through privacy pools rather than people having to go to our website, log in, use it. Um, so a feature that you could imagine is where the major wallets have like a public balance and a private balance, and you can just use one click to move in between the two. And then we come to the feature stuff.

Um, so we have spent a lot of time, uh, figuring out what you our users want and what how we can deliver that to them. And so what we want to do with privacy pools V2 is essentially create an infrastructures infrastructure stack that rivals the, uh, possibilities of a public address. And so we have introduced a whole new set of features. And so for privacy pools V1, as I said earlier, like the only thing you can do is deposit, withdraw, you pay a 0.5% fee, and it just works in your browser wallet.

And it's like, yeah, super super limited. What the goal for privacy pools V2 is is that users can keep their funds in the pool and use that as basically their their home base for their funds. And whenever they want to do something on block on a blockchain, they can essentially do that by just doing a one-time, um, interaction with with the public address and then re-deposit back into the pool. And so we have, for instance, like transfers within the pool. We also have like payment requests, which is pretty cool.

It's like Venmo, but on chain and private, so you can just like send a link, one-click pay. Um It's a nice feature. Uh we have multi-sig, safe compatibility. We have no fees, which is also like I feel like a really big game-changer because like having to pay 0.5% fee every time you go in and out of the pool just like limits the amount of times you want to do that.

Uh we have hardware wallet support, which means you can actually like uh store the CK secrets that you have on your ledger itself. Um which is like a really cool innovation. Uh we have stealth addresses and intents as well. So, where you what you could imagine is that like, "Okay, I have some ETH in privacy pools. Uh I want to open an Aave position."

Uh you could essentially just like use a one-time stealth address to do that, withdraw the funds, open your position, redeposit them back in. Um and through intents, you can do stuff like swaps, uh which would then take funds out and redeposit them back in for you. And so, here you can see that kind of like the functionality and what we are hoping users are are going to see as well is that it is there is basically no reason to take your funds out of the pool once they're in because everything that you'd want to do is already uh possible privately on mainnet Ethereum, which is um yeah, a cool innovation. Uh again, yeah, this is the the Venmo-style payments. So, um I'd like to show you guys a demo, but it didn't really work.

So, uh what you can essentially just do is is create like um a link that you can send to your friends, and they can pay that for you. And we also have the possibility now to facilitate crypto card payments. Um I gave a talk about this back in Can, where essentially through five June query prompts, um I was able to identify every single Gnosis Pay user and create like an economical behavioral profile of them because their entire like spend and profile and like everything they do with their card is public on chain. Not great. And so we built Privacy Pools V2 as well to facilitate like private card payments.

We already have some integrations live. We have Haven Network who is a card issuer that that now uses Privacy Pools for like all their transactions in stealth. And so we feel like Privacy Pools is basically built for like neo bank infrastructure and that you can basically do anything that you'd want to do with your public account privately. Okay, so there's essentially three ways that you can build on Privacy Pools V2. The as I think the easiest most low hanging fruit is build what already exists in public but privately.

Um you can integrate it into the your existing infrastructure and so it's super easy for these three different SDKs that I mentioned to embed your flows through Privacy Pools and then give your users privacy without them having to worry about it whatsoever. There's no sign up, there's no seed phrases, there's no technical knowledge barrier whatsoever. And you can also build on the ASP. So if you are developing a privacy protocol yourself, you might have issues with compliance. We're very happy to help you out with that and we are very happy to provide you with ASP services.

So yeah, the takeaway here is is that privacy isn't a feature you add, it's a default you restore. In Web2 like as much as I hate it, everything was private by default and it's actually crazy that that wasn't the case on blockchains. We're restoring that default, we're We're back the privacy in our financial transactions. Yeah, thank you for listening. Thank you for coming.

[applause]