Privacy Pools v2: towards privacy by default - Mike McCabe

Added: 2026-05-19

[00:00:12]Okay. Hello everyone and uh welcome to my talk. I'm very happy to see uh so many of you here. So I was going to talk about privacy pools v2 and how that moves um yeah how that moves us towards uh privacy by default. Uh, but I already gave that talk a little bit at ICC and so I decided to uh switch things up a little bit and talk about some other stuff and segue into privacy pools v2 eventually. Um, so yeah, I want to talk about privacy as infrastructure for digital sovereignty. And the reason I want to talk about that is basically what we've been experiencing this year.

[00:00:50]Uh, 2026 I kind of dub it as the cyber apocalypse. It is like the year everything got hacked, right? like this is just a very small snapshot and I included most of these exploits because in like three of these my KYC documents got leaked. So it's it's been a disaster honestly and like there is not a single day that goes by without um another major data breach and us all scrambling uh to see if we're uh exploited.

[00:01:20]Um, so obviously not a great situation to be in.

[00:01:26]And what we can kind of like conclude from this or what my thesis here is a little bit is that the honeypot doctrine is dead. Um, every centralized database is now essentially a strategic liability. I don't know if any of you guys have used any of these AI red team uh kits.

[00:01:48]Obviously, don't do that. It's legal.

[00:01:50]But they are insane and they are it's truly mind-blowing the capabilities that you can get from just like essentially doing free prompts and you're, you know, able to take down the FBI internal security um system. It's ridiculous how good AI has gotten. And it's like we're all monkeys with machine guns now. Uh this is the reality we live in. And you know the crazy thing about it is this is is that some of these hackers doing this developing these red team kits are nation states. Um the internet that we live in today is not the same internet as it was before. We are living in an incredibly hostile environment and uh we should act accordingly. And so my thesis here is the only winning move is just simply not to collect any data that is not necessar necessary. Uh data minimization is I think the only way forward and so that brings us well we're here in Prague that brings us to Europe.

[00:02:49]Uh Europe is facing a big dilemma. They uh want to get rid of big tech. They want to become sovereign in the digital space whatever that means. Um and they have a few options here right? Uh, so the first thing they can do is like escape big tech. And I think this has been a very funny discourse to follow. I don't know if any of you guys have seen like AWS Frankfurt or Azure Germany and this then being branded as European sovereignty. I mean, this is just a turbo LAR in my opinion. Um, this obviously doesn't replace anything as long as the parent company is still a US-based entity. Um then the second strategy that we've been uh seeing and this has already been happening for a few years right trying to replace big tech and so we've seen initiatives like Gaia X which had all the best intentions and I think has done a bunch of good work but reality just points out that outb building a hyperscaler like like your Google's like your AWS's is just incredibly difficult and by pursuing that strategy all you're doing is just like recreating these same data silos now in a different system, right? So I don't really think this is a very sustainable way forward either. Uh I think the only realistic option here that we have is minimizing the target.

[00:04:16]Um don't collect what you can't defend.

[00:04:19]I think that ZK proofs are are truly revolutionary and are going to play a major major part in in the future of the internet basically. And we are already seeing that, right? Like the EU is already shipping ZK. I don't know if any of you have been following like the new EU digital identity wallet, but they are literally like uh showing us examples of circuits coded in noir like they are aware of what we're building and they are adopting it which is like really cool to see. um we have regulation that now specifically recommends using ZK proofs for identity frameworks. So you can actually see that this shift in mindset is happening not only here in hacker spaces but also on the bigger stage which I think is incredibly encouraging to be honest. Um and yeah, where does all that all of that come from, right? This innovation like crypto is the ZK frontier. Um we have been building this kind of stuff for 10 plus years, right? Specifically Ccash obviously. Uh and the EU is now kind of like dipping their toes into identity. But we have come so much further than that already. Like we have general purpose CK compute on Aztec now.

[00:05:38]like we have really cool tech actually and it's really cool to see that being recognized now and I think uh it's going to be adopted way way more in the future and what is important about that is also that we as an industry capitalize on that a little bit right like what we've seen now in in recent kind of like months and years is that big companies like Google uh Oracle uh AWS have come in literally literally taken our good and honest open- source code and taken it for themselves and started implementing that in their own system which sucks. Um so we got to be a little bit smart about that and and see how we prevent um our systems our code essentially from being captured by by big tech.

[00:06:27]Um yeah, so I'm here with uh handing out handing out all these hats to everyone.

[00:06:33]I hope you guys all have one. If you don't yet, uh come see me after my talk.

[00:06:36]I'm I'm happy to give them to you. Uh but what I'm spreading, right, I I call myself chief propagandist. What what I'm trying to say is privacy is normal. And when I give a hat to someone, they're like, "Oh, sure. I agree with that.

[00:06:48]Privacy is normal. Of course it is." Um but if we actually take a look at the data, privacy is anything but normal on chain. Uh I have a statistic here which is just mind-blowing to me. Less than 0.01% 01% of activity on the fairium is currently using privacy tooling. We are talking about our financial data here. This is ridiculous. Um so yeah, in a nutshell that is what we as privacy pools here are trying to solve. Um you know I it's 0.01% here but in reality it's like 0.00003% or so. I mean we're talking about less than 1K daily transactions over like the 2.6 million that happen on Ethereum, right? like the number is totally negligible. Privacy is anything bought normal in Ethereum currently. Um, and our goal is to change that. So, our northstar for privacy pools is like get this number up, pump it. Uh, we want to get it to like 0.1% first, which is already like I think a thousandx or so increase than what we have now. And then we want to get it to like 0.5% 1%. And we think we have the technology here now to to kind of do that and bring that to to Ethereum.

[00:08:04]Um, so why is this number so low? Um, I mean it's hard to give like an exact reason for that, right? But I feel like in general there are there are three big risks kind of like blocking the adoption of privacy on Ethereum currently. First one is obviously like the compliance risk, right? And this has been hanging over the space for a while with like tornado cash getting sanctioned, the oh privacy is for criminals kind of like narrative. Um it's done a lot of damage to to privacy in general. And um it's it's definitely put space back a number of years. I mean I remember a few years ago people were scared to even use the words ZK and privacy together. And we've come to a different stage now where we're changing things and things are getting better. So um yeah, I'll talk about that more, but yeah, then we have the integration friction as well. So privacy is a separate app. I don't know how many of you guys have used privacy tooling here, but it is complicated, not going to lie. And I say that like as a privacy protocol developer, right? Like I wish I could make it a lot easier. Um and we think we found a way to do that.

[00:09:12]And then the the third thing is just like there is a huge UX gap here. If you look at like the the things you can do with your regular EOA or smart account on on Ethereum with just like a regular address, you can do so much more than a privacy protocol. And it is just it is just like we're not meeting the user demands from what is required for for uh a protocol. And so with privacy pool 2, we're essentially trying to change all these things. Obviously like we have our ASP um which is basically our gatekeeper whenever someone deposits into privacy pools. uh we we check the source of funds and if they're associated with like any hacks or scams or whatever uh any illicit activity basically we uh don't allow them into the pool um they can only withdraw back to their depositing address and um this prevents the pool of being tainted which is already I think a great step in the right direction. I think this largely decreases the risk that uh we experience as privacy builders. If you look at like new privacy protocols now, I think that compliant privacy is is basically the default. Uh this is what everyone seems to be shipping. So I feel like that thesis is honestly largely validated.

[00:10:25]And then we're also adding viewing keys to V2, which is just like a useful thing if you want to if you I don't know get probed by your tax authorities and they're like, "Hey, you use a privacy protocol. What the hell?" You can just give them your fing viewing key and uh that solves that.

[00:10:40]Uh second thing is of course the integration gap. So standalone privacy protocols are complicated, difficult and expensive. Uh we're trying to solve that and I think we're not the only ones here. Uh I want to give a big shout out to Kohaku which I think is going to be a major major gamecher. Uh if you don't know what Kohaku is, it is essentially an SDK for wallets. Any wallet provider can just implement this once it's released which I think is in a few months. uh and that's going to allow uh these these users to just like shield their balances with one click instead of having to go through this entire friction of like having a separate seed phrase these kind of things. Um so yeah very excited for that and I think that is going to really help us pump these numbers. Um and then 2030 plus long-term uh we have of course straw map confirming uh native privacy by default.

[00:11:33]uh extremely excited by it uh for it but we still have a long way until 2030. Um so yeah that's our northstar.

[00:11:43]Then the feature completeness I think this is really like a major thing for for privacy pools v2. This is basically the reason why we built this. Um, our current flow with privacy pools v1 is like so so basic, right? Like you can deposit, you can withdraw, cost you a 0.5% fee, and it's like browser wallet only. It's super limited. And what our goal for privacy pools v2 is to essentially create a feature complete privacy protocol. And so what we're going to have is a shielded pool with transfers and requests in there. So, one one functionality that I'm going to show you guys in a little bit that I think is really cool is like Venmo style payments, completely private onchain.

[00:12:23]Uh, we're going to have multi and safe compatibility. We're going to have unifi unified accounts. We're going to get rid of fees, which is obviously great.

[00:12:31]Hardware wallet support, stealth addresses, intense compatibility, builder codes. Um yeah, we really made an effort to basically ensure that from your privacy pools uh account you can do anything that you would like to do on mainet. So you could open like from your private balance you could open a position on a using a stealth address, deposit that back into privacy pools and manage it from your private balance.

[00:12:59]Um yeah, one thing I really want to uh talk about with you guys is come build this with us. Uh the reason for that is is that like one of our goals for privacy pools v2 would be essentially to just completely get rid of our front end and have privacy pools be embedded within existing infrastructure. So our SDK is a work of art. Um, we made it specifically so that it is super easy to integrate into whatever system you already have.

[00:13:31]And that is really our goal for this embed this in existing infrastructure rather than have it be a standalone um application. And so one quick shout out that I want to give if you're curious and uh can't wait for Kohaku. Cloaked is a wallet founded by uh one of the optimism co-founders.

[00:13:51]uh they have already essentially uh built what Kohaku is building in their in their wallet here. If you want to really experience how seamless privacy can be, I really recommend you guys check them out. They have built a great product and um yeah, we're big fans. So that's one thing I would like to uh tell you guys. Then uh what I want to do for the rest of this presentation is I see I don't have so much time left but I want to just demo you guys some things from privacy pools itself and specifically how easy it is to build stuff on privacy pools. So maybe one thing to say here is I'm not dev I am not technical. Um so I've basically just vioded everything that you see here apart from this main UI that we have. So this is the the privacy pools v2 UI. Uh, one feature that I want to show you guys here is the request feature. So, I can essentially just create a payment request here.

[00:14:48]Select Eve, create the request, copy the link. I can now send this link to anyone.

[00:14:56]And let me just open it in another tab here.

[00:15:04]>> I can confirm. And it gives me two options here. So, I can pay directly from my wallet. And then it's just like a deposit into privacy pools on behalf of the user that requested it. Or what I can do is I can pay through a private transfer which is what I'm going to do.

[00:15:19]Uh and so this is a transfer happening within privacy pools itself. What is cool about it is that it is um uh facilitated by the relayer meaning that the recipient doesn't know who which address sent the money either. So it's like a double blindsided payment request. Um let's see how we're doing.

[00:15:37]We are submitting it through the relayer.

[00:15:41]This should be done any moment now.

[00:15:45]Here we go. And I can go into the explorer.

[00:15:50]>> And you can see here that like the address I'm connected with is ox98, but the address that submitted this is ox4ba, which is the relayer. And so the private payment request got paid without revealing any information about the request. So, no amounts, no uh sender or recipient. The only thing we know is something happened.

[00:16:12]Um, another thing that I think is pretty cool that I just built. Uh, is the payroll. So, uh, we think that payroll is obviously a great use case for privacy. Uh, I just vive coded this last week. It's a super simple front end that runs on the privacy pools SDK. You can see here I have four recipients that I would like to pay from my treasury that I have here. Uh essentially what I can just do is run cycle here. I select how much I want to pay these people and I press confirm. Here you can see that we're in the discovery phase. Um we're preparing the transfers now.

[00:16:58]And this takes a little bit of time.

[00:17:01]That's just simply because we're working with CK proofs and uh it takes it some time to to process these uh the preparing uh works a little bit long but it's submitting it now. Also what we have to do here is uh we have to create like two different proofs because we're spending one ETH proof and then receiving another one back and we're spending one USDC proof and then receiving another one back. So we essentially have to wait for two cycles to complete here. bought. I think that it should happen any moment now. Yeah, I can feel my phone vibrating. So, I know the transfers are coming through.

[00:17:41]And there we go. Two of them are settled.

[00:17:46]The other ones are coming through right now.

[00:17:51]And the cool thing about this is is that we have builder codes enabled. So you can basically set your own fees if you were to build a front end like this. And that means that you can essentially just like build any sort of front end on privacy pools and monetize that front end yourself.

[00:18:08]Uh one last thing that I want to show you just to see that there's like a real variety of use cases here. Uh uh this is just an order book I've coded.

[00:18:20]I will start it from here. And what I can do is create a new order.

[00:18:27]I will do it for a very small amount, just 0.01 if and I'm going to submit this sell order.

[00:18:37]And then I'm going to go back here.

[00:18:40]And I'm going to go for my other to my other account and I'm going to fill this order.

[00:18:48]And here we can see the proofs are being generated.

[00:18:58]And here we can see exactly what's happening. So now we're first creating the maker transaction.

[00:19:04]Uh that's done. Now we're creating the taker transaction.

[00:19:13]And that should be done in a very short time as well. And yeah, what my goal for this is is just to show you guys it is super super easy to build on privacy pools and you can monetize it. You can build a whole different variety of use cases. Basically everything you can build on uh a public blockchain now. You can also build privately using the privacy pools SDK. Um it's a really cool tech. We're releasing it here. You see it completed. So I have now swapped this within the privacy pool. uh we can check on epher scan and indeed we see a transaction has been completed uh on the one end and on the other hand here as well and again you can see that like the address that um we we use to actually fill these orders is a different address from the address that we see here because we're using a real layer to facilitate this so no one has any idea who is interacting with who um yeah so I really invite you guys to come build on privacy pools uh our our test net is going live probably tomorrow, maybe the day after. Uh I've made this promise before. We had some delays unfortunately, but we're really ready for it now. Uh yeah, thanks for thanks for coming to my talk and listening to me. Uh this is uh what we're building and we really hope that you want to come build it with us. Thank you.

[00:20:39]>> Any questions?

[00:20:47]The core devs are talking about frame transactions potentially going in the Hagod upgrade which could they say get rid of relayers. How do you how do you kind of see that happening? Do you see that happening? How quickly could privacy pools do such a thing if they wanted to?

[00:21:02]>> Uh yeah. So um we've looked into this and I think frame in in in general is like a cool concept, right? Um, I was actually made aware of of this yesterday and luckily did a deep dive just in time. Uh, we we've looked into uh replacing the re layer and we've looked into pay masters as well. Um, I think it doesn't have to be as black and white as like replacing the relayer entirely. I think a relayer has its use cases like especially for things like payment requests where you want these to be double double blind. Um we are we are not hellbent on having relayers. We want to offer the best user experience to to our customers really. Um we're doing it with relayers now. If that best kind of like uh practices change in the future then we will change with that.

[00:22:02]Do you currently run your own relays? Do you plan to have some decentralized network of relays? How do you think about this?

[00:22:09]>> Uh we currently do not run a relayer. Uh there are third parties that run a relayer. If you want to run a relayer, that is totally fine by us. Um what's cool is that we have basically some like uh variables that like a relayer needs to adhere to in order to become like uh a relayer in a privacy pool system. So like um there are other privacy protocols where relayers um have the opportunity and sometimes do misbehave uh which sucks because you have to put a lot of trust into these entities. Uh what we've designed priv with privacy pools is a framework that if you want to be a relayer you have to stick to these rules otherwise you can't. Um and so we basically mandate good behavior by the relayer. Uh so that reduces the trust assumption that you have to place in there. Uh we don't run a relayer yourself, but if you would like to do so, uh be my guest.