Loading summary
Transcript178 lines
- [00:01]
Unchained Promo/Ad Host
Hey all, before we begin, I've got some exciting news to share. We've been working on something a little wild behind the scenes. It's called Unchained On Air, a revamped live stream series and podcast feed that takes you way beyond the headlines. It features sharp, maybe even controversial takes on major events and the kind of on chain intel that never makes it to your feed. Way more shows way more often, each one laser focused on a different slice of crypto and finance. First up is Dex in the city where the wallets are cold and the takes are hot, with Jesse Brooks, Katherine Kirkpatrick, Boz and V. Lee, three powerhouse lawyers gathering to dish about the latest. From defi enforcement to token regulation and everything in between it livestreams every Tuesday at 12pm ET. Second is uneasy money because what happens on Chain never stays on Chain with Luca Netz, Kane Warrick and Taylor Monahan, three OG DeFi builders unpacking everything happening on Chain, from tokenomics to daos, from hacks to yields. It airs Wednesdays at 3pm ET. And finally, bits and the Interview, an addition to our group chat show in which our Executive editor Stephen Ehrlich takes you deeper with one on one conversations. This streams on Thursdays at 12pm ET. To catch the live streams, follow Unchained on x, subscribe on YouTube or find us on your favorite streaming platform now. And don't forget to hit the bell icon so you never miss a show. And if you can't make the live stream, these episodes will show up in your podcast feed the very next day. Thanks as always for your support.
- [01:46]
Joe Andrews
We've been running kind of a test net for this since May and had 10,000 plus nodes on the testnet. But doing it on Ethereum mainnet with real economics was like the final kind of like piece of the puzzle. So we're going to run that. Well, they're going to run it for the next kind of two, three months and then early next year they'll turn on transactions and then the whole network will be live. Private World Computer will be running and you'll have privacy on Ethereum, which is kind of what we've been waiting for.
- [02:12]
Zach Williamson
Longer term, we want ASEC to be the global exchange and settlements platform for the world's assets. Not digital assets, but all assets.
- [02:23]
Laura Shin
Hey everyone, welcome to Unchained, your no.
- [02:25]
Unchained Promo/Ad Host
Hype resource for all things crypto.
- [02:27]
Laura Shin
I'm your host Laura Shin. Before we get started, a quick reminder. Nothing here on Unchained is an investment advice. This show is for informational and entertainment purposes only. And my guests and I may hold assets discussed on the show. For more disclosures, visit unchained crypto.com Today's guests are Zach Williamson, co founder of Aztec Network and Joe Andrews, co founder of Aztec Network. Both of them, both of them are co founders. Welcome Zach and Joe.
- [02:54]
Zach Williamson
Thank you.
- [02:55]
Joe Andrews
Thank you so much for having us.
- [02:57]
Unchained Promo/Ad Host
Are you a builder who needs to add on chain trading to your product? The Uniswap Trading API from Uniswap Labs offers plug and play access to some of the deepest liquidity in crypto. Its on chain execution at an enterprise level. More liquidity, less complexity. Visit hub.uniswap.org to learn more.
- [03:16]
Laura Shin
So Aztec just launched ignition chain. Congratulations. Tell me about it. And Zach, why don't you start?
- [03:22]
Zach Williamson
All right, so this has been something seven years in the making. So the ignition chain is the production of Astic blocks on the Ethereum mainnet. I should probably say what astic is.
- [03:34]
Joe Andrews
Right.
- [03:35]
Zach Williamson
So we are fully private, fully decentralized layer two on Ethereum where we want to take the world computer architecture that Ethereum created and add private information into it so that you can do private identities, private data and have smart contract smart contracts that reason about that. Effectively a private world computer. And Ignition is a major part of our rollout process where we have fully decentralized block production. And maybe Joe can talk more about that.
- [04:03]
Joe Andrews
Yeah, so kind of launching decentralized, we wanted to do it in stages. And so the first stage of this is what we call ignition. It's a bit like a Ethereum's beacon chain. And so we've deployed the execution environment without transactions on Ethereum mainnet. The gas limits effectively zero. And what it really is is the governance system behind Aztec running on Ethereum. So I think 1am on Wednesday nights we had 500 different kind of sequencers around the world stake their tokens and start block production. We also had four different proving entities, it's published on this to come and prove the first blocks. And the goal is to just ensure that block production works in a decentralized context. On Ethereum mainnet we've been running kind of a test net for this since May and had 10,000 plus nodes on the testnet. But doing it on Ethereum mainnet with real economics was like the final kind of like piece of the puzzle. So we're going to run that. Well, they're going to run it for the next kind of two, three months and then early next year they'll turn on transactions and then the whole network will be Live private world computer will be running and you'll have privacy on Ethereum, which is kind of what we've all been waiting for.
- [05:10]
Laura Shin
Wow, that's really cool. So explain what kinds of problems you were trying to solve with ignition chain.
- [05:17]
Joe Andrews
I think it's one of the first Times that an L2 is fully decentralized. And so some people kind of in the L2 space have like pushed different aspects of decentralization. But for us, decentralization is kind of, I think there's three parts to it. There's ownership of the network, like who owns and controls the network and its governance. There's who can produce blocks normally sequencing and then there's decentralized proving. And so we've run all three of those with kind of the ignition chain. And as a result, when the execution environment lives, it will be a goes live, it'll be a stage two roll up the stage two L2, which is something we care very deeply about. I think the main goal is kind of testing that out and making sure it works as intended before users can actually send transactions. And so that's the main goal. And also just bootstrapping a staking set. On Ethereum, other chains have kind of like pushed the boundaries in one of those directions, but it's rare to see kind of all three of those be hit in kind of one release.
- [06:19]
Laura Shin
Do you want to add anything on that?
- [06:22]
Zach Williamson
Yes. Well, it's pretty concise. I think for us the incentives line up perfectly to do this full decentralization. You normally have this great tension in L2s where you want to launch quickly, you want to get some traction, and then you're like, oh, I'll decentralize later. But if you're a transparent layer too, and you're selling undifferentiated block space, what decentralization means is your transactions become slower and more expensive. And so there's this massive tension. Whereas with Aztec, because our value proposition is privacy, it's very important for us that our network is genuinely neutral and it's permissionless and that, you know, we don't have centralized sequencers because then that there'll be huge pressure from governments, from agencies to put back doors into those. Just like for example, if the Internet ran through one, one corporate institution, there'd be backdoors. And there's going to be a backdoor in Aztec over Gemini's dead bodies. Yeah, yeah.
- [07:19]
Joe Andrews
I think I would add like the. A lot of people try and get privacy through an intermediary and so you can have like, you know, centralized Privacy, that's how it works in Web two, but on, on kind of a blockchain, we think it should be done through cryptography. And so that's why decentralization is, is a key part of that, to ensure that there is no intermediary sitting between you and the network. It's, it's a decentralized network and we use cryptography to get the privacy guarantees instead of.
- [07:48]
Laura Shin
And so, you know, as you have been hearing, you know, throughout, I would say the last six weeks, there's a lot of privacy efforts and there's a lot of interest in privacy. And the Ethereum foundation also announced that it was going to make privacy a priority. So how are your efforts with Aztec and the ignition chain different from other privacy efforts in Ethereum?
- [08:13]
Joe Andrews
Yeah, so I think maybe I could just give a higher level, like overview and then we can dive into like how, how privacy on Aztec works. And I think we realized nearly seven years ago now, maybe eight years ago now, it's been, it's been a long journey. Kind of like a fully transparent blockchain was not going to meet the needs of the real world. When we started Aztec, we were trying to put corporate debt onto Ethereum and so we were trying to build a product and it turned out that product needed strong privacy guarantees. And we kept running into this wall, talking to customers, talking to institutions around the need for privacy. And fast forward a few years and Zach, you can see it on his kind of outfit here, he wrote a paper in 2018 that was called the Asset variable.
- [09:01]
Laura Shin
Let's tell the audience. Zach, talk a little bit about your jacket.
- [09:05]
Joe Andrews
Okay.
- [09:06]
Zach Williamson
Yeah, this was our designer, Tiffany had a brainwave and made this up for me. So it's got the front page of the very first cryptographic paper I ever published in 2018 on it that we originally called Aztec and that's what spawned the company, the network name. And yeah, it was our first privacy protocol that allowed for confidential transactions on Ethereum. It was, by modern standards, it's very primitive. It wasn't a roll up. It used old school crypto primitives and these weird things called bonnet buoyancy signatures. But it did the job and it started us on our path to where we are now.
- [09:41]
Laura Shin
It's funny because I think it was in 2018, my Halloween costume, because it was 2018, the 10th anniversary of the bitcoin white paper. I was actually the bitcoin white paper for Halloween that year. And at the party I went to, literally only one person understood it. Others were like, are you bitcoin And I was like, I'm not bitcoin, but close. And I had laminated front pages of the bitcoin white paper on my shirt attached to my body. And people couldn't figure it out. And then one guy, one guy was like, thank you so much.
- [10:14]
Joe Andrews
So.
- [10:14]
Laura Shin
But you figured it out.
- [10:16]
Zach Williamson
But anyway, you're too early.
- [10:18]
Laura Shin
But yeah, so you were referencing. Yeah, keep going.
- [10:21]
Joe Andrews
So I think like, what I was trying to get to was like we ran into this problem of needing privacy to get real world adoption. And I think the reason that there's been so much kind of, I guess, attention on privacy coins, zcash, all of these other things over the last kind of, probably last year, actually not just six weeks, is because blockchains are rubbing up to the real world a lot more than they were kind of like three years ago. Like we're getting people to use them. And every time you go and speak to an institution or real world users, the same question comes back, which is like, well, who can see what? Who can see my data? Is it private? And so I think that's turned a lot of focus onto needing privacy. Blockchains have come on a long way since kind of like 2018. They're a lot cheaper, they have like a lot more features, but they still don't have this feature that we've been building, privacy. And so I think that's why there's a lot more attention on it these days, just because we're getting into the real world and.
- [11:14]
Laura Shin
But how would you differentiate what you guys are doing versus something like a zcash or Tornado cache or any of the other popular.
- [11:21]
Zach Williamson
Yeah, yeah. So it's all about private composability. So most of the privacy efforts that we're seeing are very application level, application layer specific projects like Tornado Cash or like Kohaku, which is like this suite of privacy tools on top of Ethereum. And I think these things are very valuable. I think it's fantastic that Ethereum is adding privacy as one of its core values, because I think it needs to. But if you're trying to retrofit privacy at the app player on top of a public chain, you're going to run into this problem of composability, because let's say I want to build a private app on Ethereum, I have my own ZK toolchain, right? I've got my own like logic for making proofs and handling private state, because that doesn't happen at the protocol layer. And now if I want to build another app that interfaces with that app. Well, now I have my own toolchain that I need to build as well as the other toolchain from this other app I need to incorporate. And this problem grows and grows and grows and grows and grows. If you just want to have contracts talking to contracts talking to contracts. And so this is what we saw in 2018, because that's how we started, we started going, let's just add, let's add privacy to Ethereum, let's build that toolkit. And then we realized, wait, the semantics of privacy are so different that to create usable abstraction layers, you really need to actually build things from the ground up, which is what we've done with Astec. You need to be able to have a platform where you can write a private smart contract with private storage, where you can just pull another smart contract which has its own private state, which can pull another smart contract and random people arrive at these contracts. You don't have to trust them and it just works out of the box. We have this thing called the Pixia private execution environment, which is basically the private part of the Aztec node that users run, where it's basically this like one stop shop to handle your privacy needs. Right? Like our smart contracts expose, you know, nice APIs that you can, that apps can call through the Pixie. So like to call these private functions to generate these proofs to send transactions and if you don't have at the protocol level, both like an understanding of private state, private state transitions and basic semantics around how to, how to, how to like call private smart contracts and the associated node software that handles it. Or if you don't have that, then you get into a complexity explosion at the application layer because you're ha. Because the app layer has to handle it and it's going to be customers going to be different for every different app and it's not interoperable and it's not composable. But with asic it is and that's the real difference.
- [13:50]
Laura Shin
So explain like walk me through a kind of transaction that's possible on Aztec or the ignition chain that isn't possible with any of the other privacy technologies.
- [14:00]
Joe Andrews
Yeah, oh yeah, yeah, I can try first. And then maybe it's good to look at kind of like zcash as an example. So zcash is like a chain that has a single asset. And what you can do on zcash is you can shield that asset and you can send it around privately. It's effectively one application on the chain. On Aztec it's like Ethereum where you can deploy any smart contract to it. But those applications have privacy and private state as a first class primitive. So we could build zcash on Aztec and we could use it, do it for any assets that would be like a simple use case. Or we could build something much more complicated and build kind of a fully, fully private dex, like a dark pool where institutions could trade with each other and the logic of those trades would be enforced by a private smart contract. So you could make sure you weren't trading with an institution who's not on the white list. And the world wouldn't be able to see what's happening except the institutions in the group. And so you can have much stronger degrees of kind of composability, programmability than a single kind of use chain that just has like one use case built into it.
- [15:06]
Laura Shin
So it sounds like it's like an Ethereum where everything is private.
- [15:10]
Joe Andrews
Yeah, exactly, yeah.
- [15:11]
Laura Shin
Like you can have like ERC20, but they're all private. You can have.
- [15:14]
Joe Andrews
It's Ethereum where we've taken privacy and made it a first class primitive within, within the network. And so Ethereum's missing that today. And so we've, that's basically been the goal of us, our network for the last years.
- [15:25]
Laura Shin
Yeah, but so what about like, you know how in zcash there are shielded transactions and then there's the public ones? Like are there these kind of hybrid type transactions on, on ASEC or is everything just private?
- [15:37]
Zach Williamson
So you have, the application developer has a choice. So you're in our smart contract programming environment. So it's a language we've written called Nor, which is based on Rust. You can create both private functions in your contract that handle private state and public functions that handle public state. And you can call the public ones from the private ones. And so it's very much up to the application developer what they want to expose, what they need to expose. So can I give an example transaction? Because I'm looking at that bit because I can just show you something that's possible. So let's say, okay, so as an entry point we have account abstraction. So the first thing that gets called is your account contract. Let's see, that validates the Google authorization sign in process for your, for your, for your Gmail account. So you're basically proving I have a Google account that's, that's you don't have a private key. That's who you are for like a hot wallet for retail. You then the next call that is made in, within that transaction is you want to Talk to a private Dex that's handling some kind of asset. So tokens that are like somewhat regulated or you know, the real world assets, they have some kind of compliance. So your Dex is now making a private call to private token contracts. And those private token contracts might say make a call out to a private NFT contracts that validates that you have been issued a credential from something like a ZK Passport contract that issues credentials based on passport proofs to approve your transaction. And so let's say it happens for both tokens, they then call back in to like return back into your. Into the private Dex smart contract which enacts the trade. And so there you have a very complicated system where you have like the account contract using Google Auth, that's one component. You have a private Dex, you have compliant token standards and you have identity stand like contracts and they've all been written by different actors, but they can all be composed together into what is from the user, for the user, an atomic singular transaction to execute a trade.
- [17:23]
Laura Shin
Wow, that's so interesting. And when you say that like meaning all within one block.
- [17:29]
Zach Williamson
Yes.
- [17:30]
Joe Andrews
Yeah.
- [17:30]
Laura Shin
Wow. And what are the block times?
- [17:33]
Joe Andrews
So it's. Yeah, it's a good question. So getting all this working has been kind of a huge like engineering effort. So at the moment that between 36 and 72 seconds. But the plan is to get them down to sub Ethereum kind of.
- [17:45]
Laura Shin
Wait, I didn't hear you, you said.
- [17:47]
Joe Andrews
70 between 36 and 72 seconds. And depending on the sequence's availability the plan by the end of 20, 26 is they'll be below an Ethereum block and we're targeting four seconds.
- [17:58]
Laura Shin
Okay. And is the reason that it's like a longer block because that there's more computation because of the privacy or like.
- [18:05]
Joe Andrews
At the moment it's just because we're, it's one of the first decentralized L2s and we want to kind of. We've done a lot of firsts with this and so we wanted to make sure everything's reliable and the sequencers could basically choose a parameter in their node and they can crank it up over time. And so that's kind of the state we're in right now. It's mostly just yeah, it's new software and if we go, you know, try and run a hundred miles an hour on day one, it will probably fall over. And so we're going to ramp it up slowly.
- [18:30]
Zach Williamson
Yeah.
- [18:31]
Laura Shin
Okay. And do you have like a goal of, is it just the goal to be the Same length as an Ethereum.
- [18:37]
Joe Andrews
Block or our kind of internal goal is four seconds. Yeah, of course, yeah, yeah. I think beyond that it's kind of like, you know, we're not playing the kind of extremely high frequency kind of like trade. Like you can always do that on an app specific chain and like.
- [18:51]
Zach Williamson
Yeah, which we will be at the.
- [18:53]
Joe Andrews
Yeah, which, which could have much faster finality, but doing that kind of like with the full decentralization of Ethereum, it's kind of harder to get much faster than that and you get kind of like diminishing trade offs below four seconds really.
- [19:06]
Laura Shin
Okay. And are transactions much more expensive because of all this cryptography, keeping it all.
- [19:11]
Zach Williamson
Private or much more expensive? I think certainly it's not going to be as cheap as purely transparent networks. Because there's more data, right? Because if your data's encrypted, there's more of it which needs to be put on blobs or. Well, right now it's Ethereum blobs, so I think it's still pretty cheap. What are the numbers that we're, we're.
- [19:33]
Joe Andrews
Totaling kind of between like 2 and 5 cents for a basic payment. And I think the majority of that, as Zach was saying, is all data going to Ethereum blobs. You kind of don't need to broadcast all that data because it's a peer to peer payment. It will be kind of maybe a fraction of a cent. The, the interesting thing about the Aztec architecture is for private transactions, all of the compute happens on a user's device. So if, if I'm proving a transaction locally with client side proving, I don't use the traditional model of asking blockchain nodes to execute for me. I take the transaction, I execute it locally on my device, then I prove that it's executed correctly and it follows the rules of Aztec and I send the proof to the mempool and all like for a purely private transaction, the only work a sequencer needs to do is verify the proof, which means that kind of we get this kind of like scaling effect around private transactions and you can actually what you're paying for in compute is a lot less than a network like Ethereum. As a result, on the public side.
- [20:36]
Laura Shin
Basically the proving is like off chain.
- [20:38]
Joe Andrews
It'S on the user's devices for private transactions.
- [20:41]
Zach Williamson
Yeah.
- [20:41]
Laura Shin
Okay. And then so the chain is more like lightweight because you just have the proof.
- [20:47]
Joe Andrews
The phone, the kind of the compute sequences are running for private transactions is more lightweight. For public transactions it's the same as Ethereum. So if you have like a transaction that's only has private functions in the smart contract. It's. It's lightweight for kind of a transaction that may do decide to do something public fee at the same time. Like let's say it's using a, a Uniswap clone and it has like a two public pools, then that would function like Ethereum and the sequencer would have to execute that portion of it to figure out what the state is. So it's, it depends on how the transactions kind of composed.
- [21:20]
Laura Shin
Okay, okay. And just out of curiosity, so like on my phone when I do that, is it using like, like I'll hear, you know this phrase trusted execution environment. Is it using like a TEE or what is it using on my phone to do?
- [21:35]
Joe Andrews
It's using all, all available cores and kind of compute to construct down a zero knowledge proof which Jack can explain probably.
- [21:42]
Zach Williamson
Yeah, yeah. So it's all using ZK proofs not to. So if you're sending a transaction on your phone, your phone will be going and like running language proof like relative to all your other stuff. It's pretty damn fast. But it's maybe like 10 seconds for transfer. But no sensitive data leaves your device. Right. You have the private execution environment, which is basically a runtime environment on your phone. It's not using any enclaves or anything. But if we've done our job correctly, it should not be possible for other applications to gain data from that from the Pixie. And so you're generating your proofs on the Pixie, you're sending your proofs out to the world, but no sensitive information leaves your device, so you have no real need for it.
- [22:25]
Laura Shin
But is it using the same part of the phone that like reads my eyes or whatever?
- [22:30]
Zach Williamson
No, because the phone doesn't have like that. That T doesn't have the kind of compute bandwidth to make ZK proofs. If you're like. If the program that you're executing requires digital signatures or it requires like sensitive credentials, then yes, that part will be. The signatures will be generated by or by the C curve enclave. But the actual proof of everything that does happen just directly on the phone's bare metal.
- [22:52]
Laura Shin
Okay.
- [22:53]
Unchained Promo/Ad Host
Hey, founders and developers, if you're looking to bring on chain trading to your product, wallet or platform, check out the new Uniswap Trading API from Uniswap Labs. It's your plug and play gateway to global on chain liquidity. No deep crypto experience required and no need to manage complex integrations or ongoing maintenance with the Uniswap Trading API.
- [23:15]
Laura Shin
And you'll get enterprise grade on chain.
- [23:17]
Unchained Promo/Ad Host
Execution, combining both on chain and off.
- [23:20]
Laura Shin
Chain sources for the most competitive prices.
- [23:22]
Unchained Promo/Ad Host
Simply put, more liquidity, less complexity. And this isn't just any API. It connects directly to the Uniswap protocol which has securely processed over $3.3 trillion in total volume with zero hacks. So stop worrying about liquidity infrastructure and focus on building your product. Get access to the same liquidity that powers billions in swaps through one powerful API. Visit hub uniswap.org to learn more.
- [23:51]
Laura Shin
So before we get into more details around the chain and the token sale and all these things, I'm curious to hear how you guys got into crypto and how you came to found Aztec.
- [24:01]
Zach Williamson
Yeah, you want to start? Long story. Okay, so everyone comes to blockchain from weird roots. There's not like you can take a blockchain degree and get a blockchain internship. So I was a, an ex academic. I was a particle physicist doing a PhD in Optus. Then I basically crashed out of that, got my Ph.D. and left because I decided I can't be an academic. It's too, it's just not for me. Became a programmer, was at a bit of a loose end. And then my brother at the time, who was an opera singer, who strangely was massively into crypto because of one of his best friends, said Zach, my best friend who works in the city, he's trying to do this fintech thing on Ethereum. He needs a technical co founder, you know how computers work. Do you want to meet him? And I'm like, and I'm not doing anything interesting right now, so yeah, sure, why not? And that's for me, that's how it started. And then Joe joined very, very quickly afterwards. Do you want to share your, your story?
- [24:54]
Joe Andrews
Yeah, I guess we both had a bit of a strange way into crypto. I was previously kind of in the food tech industry, so I was like building.
- [25:02]
Laura Shin
Wait, did you say food tech?
- [25:03]
Joe Andrews
Yeah, food tech.
- [25:04]
Laura Shin
I didn't even know there was a food tech thing.
- [25:06]
Joe Andrews
But anyway, it was like, you know, in America there was like these meal delivery services like that kind of like food tech kind of bubble. And so I was, I was doing that in kind of 2013-2017 and I actually kind of met Vitalik and Gavin in 2013. I think it was in San Francisco in the original, like Ethereum, like roadshow. I didn't pay much attention to it because, you know, it was very early back then. But I kind of followed it since then until about 2017. And when the food delivery kind of company like that, like, didn't work out anymore, I joined an incubator in London called Entrepreneur First. And Zach and Tom were on that with this kind of idea to put kind of, you know, corporate finance onto Ethereum. And I was like, okay, this is. I've been learning about Ethereum for all this time, but never jumped in. It was the right time. And so we joined that and we ran into kind of this problem of trying to put these loans on Ethereum. And as a result, privacy was the problem we actually solved. So it was kind of a roundabout way of getting in. But it's been, I think I, you know, I think about it more about, like, the reasons why we're here, not how we got here. And I think it's just based on, like, for me, it's the impact that it can have on, on the world. And so previously in my life it was like, Healthy Meals did that. And then I was like, okay, this technology is way better than Healthy Meals in terms of impact, so we can have. It was just exciting to get into it.
- [26:28]
Zach Williamson
Yeah, I mean, I remember me like running into Joe, this accelerator which we joined, because I'm like going to Tom. Like, he's a finance guy. I'm like an ex researcher, like, you know, like tech guy. And I'm like, I'm like, buddy, like, we don't know how to be entrepreneurs. We need to go to, we need to go to entrepreneur school. So it was accelerator.
- [26:44]
Joe Andrews
And then we met, we.
- [26:45]
Zach Williamson
We met Joe and Joe's like, Joe's very good at products, product design. Like, he can make things look good, he can build products. And I'm like. And I'm like, oh, my God, we need this guy. It's like, we need this guy to make, build a product. So, yeah, that's how we started collaborating.
- [26:59]
Laura Shin
I love it, by the way. I did study abroad at Oxford and I love it there.
- [27:03]
Joe Andrews
It's amazing.
- [27:04]
Zach Williamson
Amazing. Yeah. Yeah. Beautiful place.
- [27:06]
Laura Shin
Yeah. And so out of curiosity because, so I understand you were introduced to Ethereum and you were interested in privacy, but why did you decide to build it as an L2 on Ethereum? Because you could have built your own chain that was EAM compatible. I mean, there's so many other ways you could have gone about it. So, like, why the structure?
- [27:27]
Zach Williamson
It was controversial at the time because we started this L2 a long time ago and I remember, like, some of our messages going like, why are you doing this? You know, you're just, it's going to Be less valuable. You can raise way more money as an L1 than an L2. But for us, it was originally a no brainer for two reasons. One of them is play your strengths. We didn't want to build a consensus network from scratch and we didn't want to have the responsibility of maintaining that consensus network because we saw, honestly we saw a little bit what happened with zcash in that vertical because they ran into a lot of problems after they launched where they just had a lot of work they had to do to keep the zcash client up to date with Bitcoin because there was a fork of Bitcoin originally. But so all the Bitcoin upgrades and security things they had to kind of backport into zcash. But they changed everything around so much. It was not a simple process and it just sucked their time away from their core strengths, which is zero knowledge, cryptography and privacy. And I was looking at that and went, I don't want us to have to do that, let's just use Ethereum.
- [28:24]
Joe Andrews
Yeah. And I think the second part is to actually get a lot of security on your network, you have to really solve that consensus problem. We're very lucky. We have very short term consensus on the Aztec network and it's for about 20 minutes before it settles on Ethereum. And so we benefit massively from the security and the properties of Ethereum. If we had to kind of, you know, long term secure all Aztec transactions ourself, it would be a much, much harder problem to kind of build a decentralized network. And I think the second reason as well is just, you know, despite what you read on crypto Twitter, and despite people trying like, Ethereum is still like the number one programmable blockchain in terms of like developers and capital. And so like, bringing privacy to that is a much more attractive proposition than trying to rebuild that on a new chain. And no matter how good your bridges are on these kind of like ones, it's not going to be kind of as efficient to bridge as it is from ethereum to an L2. And so that's kind of the main reason we're on Ethereum.
- [29:26]
Zach Williamson
Absolutely. We didn't want to have to be like the only way you could get into Aztec is through a trusted bridge, because privacy is only useful if you have value on your network, because nobody cares how many magic beans you have. So you don't need a privacy for your magic bean network with synthetic instruments. So you need, you need assets, you need liquidity. And Ethereum is the chronological home basis in web 3. And I very firmly believe that will continue to be the case for a very long time to come. And so yeah, being directly connected into that ecosystem for us, it was extremely important.
- [30:02]
Laura Shin
Okay, yeah. I mean I can even just analogize this to like my company. It would be like, you know, if we have a subscription newsletter if instead of, you know, using one of the newsletter websites that already has that built in, if we decided to like, you know, create our own like connection to credit cards or you know, whatever. So it makes sense. So, you know, here we are at this time where Ethereum, the Ethereum foundation just made an announcement that they are going to make privacy kind of a first class priority. So I wondered like what that means for like Aztec or the ignition chain. Do you feel that it's something like that you are included in that or do you feel like it's competitive or how do you view that initiative?
- [30:45]
Zach Williamson
Yeah, it's, it's. I, I see it as very complementary. I think Ethereum has this strong need for neutrality and so whilst my, my very biased opinion they should just be like, yeah, we're going to just do bridging stuff into Aztec and get privacy that way. I feel like they will only do that if that's, if all other avenues are exhausted. So but I see them as complementary because I think that so what they're currently doing with Kohaku and with the basically like adding in privacy tool chains on top of Ethereum that they want to get integrated into wallets, it's valuable, but the spectrum of what you can achieve there is very limited because you are limited to the privacy primitives that they are creating. It's not fully programmable, it's not private composability. And the reason is because that's very hard. But we tried to do that on Ethereum 5, six years ago and we realized it's not going to work. You need to do from the ground up. And so I think that they're complimentary because what you could do on ASIC is always going to be substantially more advanced than what you could do on Ethereum with regards to privacy. I think that it's going to be a lot easier to develop for. Nevertheless, it's still very important to add basic privacy primitives into the existing Ethereum ecosystem so that everybody can enjoy privacy.
- [32:00]
Joe Andrews
Yeah, I think maybe I can add like at the moment if you bridge from a very public Ethereum into something like Aztec, that that's a data leak which like every time you come in and out of Aztec you can kind of, if you're going to a public world, you can see a lot of information. So the more private Ethereum can be made and like just the basic value transfers, it would help privacy for everyone. But I agree with Zach on kind of the point around the expressivity of what you can do. On Ethereum versus Aztec, we had, we've had three iterations of Aztec and so the first version was kind of like basic privacy. And then we had kind of like basically zcash on top of Ethereum for lots of different assets. That was called Aztec V2. And then we built a system called Aztec Connect that let you kind of bridge those funds to any Ethereum smart contract. And along the way we kind of ran into a lot of the same problems that they're facing today, which is kind of how do you, how do you kind of make composable applications work and not these like siloed applications? I think the only real way to solve that is something like Aztec. So I think for single use applications, it may get some traction and it will help people get a basic level of privacy. But for those advanced use cases and kind of like really verifying things like passport proofs and the kind of flow that Zach gave earlier around a Dex, I think it's going to be hard to do that well on Ethereum, at least not anytime soon. So it's about the timeline as well.
- [33:27]
Laura Shin
And tell me about what types of demographics you think will use Aztec or the ignition chain or what types of use cases you expect to see with it.
- [33:39]
Joe Andrews
Yeah, in terms of use cases, I think we're definitely going after kind of use cases that institutions care about, but the approach we're taking is kind of not necessary to go after institutions at this stage. We're finding kind of a lot more success with kind of teams of two or three hungry kind of like startup builders who want to go after an institutional use case and build like a competitor. And so we're seeing that kind of in the identity space. We're seeing that in kind of the Dex space and the payment space. So I think it's more kind of like hungry Ethereum builders who've faced problems and building kind of institutional use cases. At this stage, there are some like, institutional conversations happening. And I think at some point they're going to meet in the middle, like an institution will agree to a pilot and someone will have built a proof of concept and there'll be a kind of an amazing meeting in the middle.
- [34:31]
Zach Williamson
All right, babe, I'll expand on that. Like we have short, medium and long term goals here. Yeah, short term web3natives, you know, as we saw that connect 60,000 monthly active users. People want privacy on chain, particularly if you have portfolios or if you have just a lot, a lot of capital at stake where you have, you're like, you know, you have some hedge trading strategy and that doesn't really work if everybody can see how you're hedged. So the base, basically we create the basic defi stack on Aztec, you know, lending markets Dexes, but make it private. Then we also want to do basic identity primitives. So like symbol protection basically. You know, if you want to do some kind of sale or some kind of distribution where you want to make sure that this one person, one like uniqueness checks, you can do that on nastic very efficiently and very easily with things like ZK Passport, even like a ZK TLS proof, you have a Twitter account with over a hundred followers. I mean that's, you're probably not a bot. You might be, but you know, it's a lot harder to get 100 Twitter followers than it is to spin up a bot farm then. And like, whilst we're doing that in parallel, we want to absolutely incubate this basically like this tradfi disruption track where you're getting these small hungry teams to build products and services that enterprise talk about, but are trying to do in ways that in my opinion are fundamentally flawed using like decentralized third parties and private chains, like private isn't private database and disrupt that and generate social proof and create a lot of fear amongst these institutions that they're going to get disrupted and use that fare to bring them in and get them to join us instead of trying to avoid us. And then longer term, we want Aztec to be the global exchange and settlements platform for the world's assets. Not digital assets, but all assets.
- [36:21]
Laura Shin
Wow. Honestly, actually when you were talking I was like, I could imagine there being a perp Dex. Just, you know, right now we're obviously seeing the flourishing of the perp Dex. Um, well, so I did also. Oh, and just quickly, because you mentioned ZK Passport, can you just explain what that is for the audience?
- [36:39]
Zach Williamson
Oh, of course. So it's so pretty much every passport these days has a little NFC chip in it which contains a digital signature signed by your government, where the message is all your passport data. So your photograph, your date of birth, your name, your passport number, blah, blah, blah. And you know, phones these days, they have NFC scanners, so you can, you can extract that with an app. You can extract that signature. Problem is, if you use that digital signature as an identity attestation, it leaks a lot of information because it contains within it, well, your photograph, your passport number, your name, your date of birth. And so what ZK Passport does is it uses our technology to selectively disclose restricted statements. For example, you can use ZK Passport to prove that you're a citizen of a certain country or that you're not a citizen of a certain country. You can use it to prove you're not on an OFAC sanctions list where what you're what's getting what the. But where the proof is basing an on effect sanctions list. And here's a hash of hash of my passport details. And that's what the way we're using it. We're using it to get basic checks about somebody in a decentralized and strongly permissionless manner.
- [37:42]
Laura Shin
Okay, so I am so curious to ask you guys, like, what do you make of this craze in privacy all of a sudden? Like, it sort of felt like it came out of nowhere at the end of September. Like, you've been working on this for years. Like, yeah, did that surprise you? Or like, where do you think it came? Like, what were you thinking watching this whole thing?
- [38:01]
Joe Andrews
Should I start? Yeah, you start.
- [38:02]
Zach Williamson
I'm thinking. I told you. I told you. I've been going around to these conferences for seven lemon years going cooking about privacy. And I've been saying exactly the same things that everyone is excited about now. You know, you can bring in real world assets, you can bring an enterprise in institutions. You can actually do stuff in the real world. You can, you can. You get basic user protections. You stop this thing from being a surveillance dystopia. And now everyone's like, hey, Zach, a primer zeitgeist. No, no, no, no, no, no. I didn't say that. They're like, wow, you know, privacy is a big deal that you can use it to bring real world assets on chain and stop this network from becoming a surveillance dystopia. And I'm like, yes, yes, It's a good thing. I've been building this for seven years, Joe, because now we. I've. I've got. I've got. I've got you, fam. I've got. We've got you.
- [38:41]
Laura Shin
Oh, good timing for you guys.
- [38:43]
Joe Andrews
Yeah, yeah, it's. I don't know. I think it's kind of. Yeah, I kind of said it a bit earlier, but I think it's. It's because we're running into the problems and like we ran into them earlier than most people because I think we were solving a particularly niche use case. But I'm happy that more people are seeing it the same way now because it's validating and it's also just the more kind of people building in this space, the ideas develop faster and it's kind of like an explosion of new ideas, which is always helpful.
- [39:12]
Zach Williamson
I think another thing which has caused this is basically exhaustion because what you can do with a regular smart contract hasn't changed in years. You know, yes, it's faster and it's cheaper, but what you can do hasn't changed and the design space is tapped out and it's led to this fatigue. Right. Where if you're a dev coming into the space and you're like, well, what should I do? And it's like, well you can make a dex. Well, there's loads of dexs. Yeah, but you can make another one. It is like, well, the surface area for innovation is very restricted at the application level, but not with privacy. With privacy it explodes again because you can get all these trustless attestations from the real world and you can reason about like private information and stuff that you normally can't put on a blockchain and, and people are excited by this. The technology is now matured enough that people, like, people can start to see the results of this a little bit. Like you had, you know, like last year you had a noncast, we could do anonymous social media posting. You know, you've had, I think Aztec has been building on this narrative, building this narrative for a very long time. You have technologies like Noir where you can write really efficient feature rich private programs. And so previously what we would, Joe and I were going and talking, we were talking about a hypothetical world and now people are starting to see the seeds of reality creep in and they're like going, oh wow. Yeah, that's really a thing. So that's now why it's I think become a lot more mainstream.
- [40:32]
Joe Andrews
Yeah, even within the company. I think like when we first started the number of people who could write zero knowledge circuits was probably three, two or three people. And now like with NAR and these like tools getting a lot, a lot more kind of powerful, it's been kind of access to writing private applications been brought to like, like a lot more developers, like most Ethereum developers who can write smart contracts can now write private applications. And so I think that's part of why the craze is happening. And then also just like, I guess a cynical view on my side is that a lot of projects have launched and maybe they didn't get the traction that they were hoping for. They had a thousand TPS of promised throughput and they were kind of nowhere close to that. And they started looking at why and like what other features needed to be built. And they probably ran into the same, hey, we need this missing feature of privacy. And so they, they started kind of jumping on the same bandwagon. So that's a cynical view, but I think it's, it's somewhere in between real world adoption actually requiring it and the cynical view.
- [41:29]
Laura Shin
Well, I mean the reason, you know, when I mentioned the perps thing, like I was thinking about, I'm sure you saw the whole James Wynn thing and people saying, oh, you didn't follow it on Hyper Liquid.
- [41:40]
Joe Andrews
I've been busy trying to launch the ignition. Bit of a. Bit of a cave.
- [41:45]
Laura Shin
Oh, okay.
- [41:46]
Zach Williamson
It like magicate me.
- [41:48]
Laura Shin
There was this guy who, well actually he's, he's British I think, but he was trading on Hyper Liquid. But since everything is public, people were saying maybe he was being hunted because he managed to get, I think it like he started with like 13 million and he managed to get it up to like 100 million but then promptly like lost it all. Oh, and people were saying like, oh, maybe he got hunted because his positions were all public. And then recently there was another trader on there actually. Yeah, that person also left. I'm blanking on like the exact reason, but I do think there were people that were copying that person's trades again because it's all public. And then like 10, 10 happened. And so I think a bunch of those people were liquidated to that like the copy traders. So that's why I was saying like like a perps Dex, you know, because alto, what was happening was CZ was criticizing the. Criticizing HyperLiquid and, and the founder of Hyper Liquid and CZ were like having these Twitter posts back and forth and Twitter and CZ was like, oh, well, maybe I'll make a private, you know, perp Dex. Because like, you know, this public experience is not a good experience. And anyway, point is just that, yeah, I think, I think there's probably a lot of reasons, but yeah, it's amazing you guys were like so early and then kudos to you for like such amazing timing. Like people are finally interested and like you guys launched now. So I.
- [43:08]
Joe Andrews
Can I give an example on the hyperliquid one? It doesn't just apply to hyperliquid, but any kind of like defi primitive that's kind of exciting with privacy. So because the Aztec block space is differentiated, it's private. You don't need to rebuild some of these things on Aztec. You only need to build things on Aztec. If like you're changing the core primitive to require like the underlying like products require privacy. If you have an already public defi application, it's sufficient to just send your transaction through Aztec. And so like in the kind of like hyperliquid or Perk Dex example, if you send a transaction through or from Aztec to hyperliquid or lighter one of the ZK L2s, well it just looks on chain like it's come from Aztec. So you can't see the kind of surrounding positions, you can't see the portfolio because they can, they can hide that from the world. So in that scenario you could have the same trader who has a short and a long that are hedging each other and you can't link them. So you can't kind of like do the high ticket example and you don't have to rebuild everything on Aztec. So there's this kind of like interesting world where because Aztec has private block space, it's actually kind of value additive to all L2s and all of Ethereum because we're not competing for the same types of transactions. Yes, like users can put their funds on Aztec and interact with all of these chains via kind of bridging and they get a degree of privacy doing that. Some applications that like are more advanced than that, that you know, you want to put passport proofs inside them will have to be rebuilt on Aztec or like some, some parts of it will be. And it's just an exciting thing because it's. I think people haven't realized yet that like private bridging and private intents are going to be this kind of way of getting privacy in the short term on a lot of public applications.
- [44:47]
Zach Williamson
I mean we've seen near like really like make people start aware of this with nearly and zcash. But to really expand on this thing to get to the essence of it. Most L2s are parasitic to their L1s. Right. Because they're just trying to create their own liquidity moats and recreating the same ecosystem. ASIC is very differentiated in that it's very positive sum to interact interoperate between astic and other L2s. Other L1s. Because there is value in bridging not assets, but information into and out of Aztec. As in if you need to, if your app needs Some sort of identity check or it needs to check against civil resistance, or it needs to perform some validation on information that you can't broadcast publicly. Well, if you're running that app on Solana or on base, you can send a request into asset going, hey, can you check this credential? And then the asset contract will just return yes or no. And so you don't disclose the sensitive data, but you get the attestation you need, you get the intent that you need. But in that example like the assets and the actual financial activity is still stay on the transparent L2 or the transparent L1. And so it's very positive sum. Like we get the asset gets some fees because you've done a credentials check on Aztec. But we're not siphoning liquidity. So one thing we really want to genuinely execute on over the next 12 months is to make Aztec a very canonical privacy shield for all of Web3. Not just ourselves, not just Ethereum, but all of Web3 with efficient seamless bridging.
- [46:19]
Laura Shin
And just understand how that would work, that you know, having something on base and then you like use some part of Aztec in order to, to confirm something. Yeah, would that use the Ethereum interop layer? Because I just did an episode on that and I don't know like currently, like would you have to bridge or.
- [46:38]
Zach Williamson
Like you'd have to bridge as in like right now we're just, we are. When it comes to like the cross chain interoperability, we want to focus on simplicity. Um, so like for example, we're partnering with Wormhole to, to to do that. And I think like the whole Ethereum interrupt layer is like, it's something we could plug into. It's just that like we need to be very mindful about things like the cost the user is paying and like complexities around executing that transaction.
- [47:01]
Laura Shin
Okay, yeah.
- [47:02]
Joe Andrews
So at the moment it's using like it's using a relayer to. So use a sensor transaction on Aztec. You can't see who sent it, but there's an intent in there that says someone on Aztec would like to swap one eth for some DAI on base. And a relayer comes along, picks that up, fills the order and brings the assets back to Aztec. And so that's kind of like one flow. So it's done with relays. But you can get pretty good kind of security guarantees around those. Because we were a ZK Rollock, you can verify what happened on at least any L2. If you're bridging out of Ethereum, then you have Some additional trust assumptions that you have to kind of deal with, but it's is mostly done with relayers. The exciting part about why it's different to normal bridging is just you can't see who sent the original transaction. And so that's kind of the kind of privacy angle that we're adding.
- [47:50]
Laura Shin
Okay, so now let's talk about the decentralization aspect because people have been bemoaning for a very long time that we have all these L2s and they're not fully decentralized. And you guys are launching decentralized or you launched a chain that is decentralized from day one. What did it take to do that? And I'd be curious to hear you opine a little bit on why it is that there aren't many others.
- [48:18]
Joe Andrews
I guess, yeah, the cynical view is like most public kind of L2s, they don't really have an incentive to decentralize. Like they have a centralized sequencer that prints somewhere between 40 and $150 million of fees a year. And so it's very easy to kind of appease the Ethereum ecosystem by saying, hey, we'll decentralize later. And then really they get shackled to kind of these cash flows and the reality is later never happens. And so I think that exists because they can put themselves in a privileged position. In kind of the case of like a privacy chain, we didn't want to be in that position because we don't want to kind of be in control of users data and kind of be in that position. It has to be a neutral privacy layer like we have on the Internet. And so that kind of forced our hand a little bit to kind of solve the problem earlier. I think like actually doing it, like it took a lot of kind of different research. We worked with kind of like our community quite a lot through rfp processes and RFCs. And I guess, yeah, we just kind of iterated till we had a design that worked. I will say that we like, we're not immune from the kind of like we'll do it later track. Our previous products that we had were not decentralized. We had a version called Aztec Connect and it was centralized and we learned a lot of lessons kind of building that that we kind of made some mistakes that we didn't have to make in this version. So we kind of learned from that how to do it. But yeah, I do think that Ethereum as a whole does need to kind of move away from like L1s that are L2s that are basically just databases at this point and actually start decentralizing the L2s and getting more of them to stage two.
- [49:59]
Laura Shin
And so explain just on Aztec like how the sequencing happens, just like the nuts and bolts of how you decentralized it.
- [50:06]
Joe Andrews
Yeah, it's very similar to the Ethereum design. So stakers have to stake 200,000 Aztec tokens. That's the minimum staking requirement. And they then get randomly selected to be on a committee. Committees are 24 at the moment and they'll expand to 48 next year. When you're on the committee you kind of will be elected to produce a block in your slot and other committee members have to check that the data that's needed to prove your block is available. And so we have these rotating committees who kind of work together to effectively give a very fast chain of pre confirmations of what could be proven on Ethereum. And they commit that chain to Ethereum every, you know, 36 seconds as we said earlier. And that will come down in the future. There's then kind of this chain of like pending blocks on Ethereum that have been committed to and you can kind of think of them as like being secured by the Aztec staking set, but they haven't yet been validated by Ethereum. And so then provers come along and they look at that chain in blocks of 32, we call those an epoch. And they will, once the book has ended, they'll collect all the data that's needed to actually create a roll up proof, like a ZK proof that aggregates all the transactions in every single one of those blocks. And they'll kind of crunch that proof down into one single proof and send it to Ethereum that will be like verified. And so that's a validity proof that verifies all the transactions in all of the blocks in every, in the epoch. And once Ethereum says tick then it's finalized on Ethereum. And so we've kind of had to decentralize both the sequencer selection side of it and the prover side of it as well to allow kind of both of those to kind of work together and under the hood kind of the network's economics power that and make sure that there's always sequencers there, improvers there to kind of show up and do their duties on the network.
- [52:02]
Laura Shin
Okay, so I think this brings us then to the token sale. Explain like yeah, just maybe talk a little bit more about how the token powers all that.
- [52:11]
Joe Andrews
Yeah.
- [52:12]
Laura Shin
And then we can talk about the sale.
- [52:13]
Joe Andrews
Yeah, so I think when I Think of like decentralization, I think I talked a little bit about it earlier, but it's decentralizing, like many things. And like the one that most people kind of forget to decentralize is ownership of the network. And so ultimately kind of ownership and control. And so what the Aztec token is used for is it's used for governance of the network. It's the governance token. Like you can make changes to the network, you can make changes to the fee schedule, how rewards work, how the, what the network is. Sequencers have to stake the token to participate in block production. They earn their rewards and present their rewards in the token. And then finally users pay fees in the token. And the fee model is very similar to Ethereum. We use the same kind of EIP 1559 style mechanics. And so if a block is like over full of transactions, a portion of the token is burnt. If it's underfilled with transactions, then sequencers get some of the fees, but the gas price effectively on Aztec will come down. And so yeah, it's kind of used for three things in the network. It's staking fees and governance.
- [53:20]
Laura Shin
Okay, and you decided to do an ICO rather than an airdrop. Can you talk a little bit about why you went?
- [53:27]
Joe Andrews
Controversial decisions, but I think for a proof of stake network, it's very hard to gift someone their stake. So if you kind of, if all of the sequencers running Aztec were kind of gifted their stake, they would have no skin in the game. And so, you know, if the network was running into issues late at night or, you know, there was bugs, it's very easy for them to just be like, ah, I'm going to turn my machine off because like, I got it for free, like I don't have any skin in the game. And, and also for stakers as well, they have to do their duties well. Like there's things they can do wrong and they could be slashed. And if you have no skin in the game, slashing doesn't really affect you. It's like you're wearing a suit of armor. You don't care like if you get into a sword fight as much. So for those reasons, like, we wanted to create a process that decentralized ownership as much as we could, but without kind of giving ownership away for free so that people didn't care about it. And I think being in the space for like seven, eight years now, we've seen a lot of launches happen and airdrops can be very short term. Like if you look at the data around who holds tokens after an airdrop or who is still in that community. It's, it's great for creating initial hype and initial kind of marketing around a project, but it's not, it's not very good for long term aligned community holders. And so building a decentralized network at the start with an airdrop would probably result in all of our staking, set churning within three to six months and our community kind of suffering the result of that. So we wanted to try a different approach where we're kind of changing the norms in the industry a bit and we wanted our community to kind of grow with us. And so kind of fair ICO was the or token sale was the approach we took.
- [55:15]
Zach Williamson
Yeah.
- [55:15]
Laura Shin
And you're also using this Uniswap Lab's newly launched Continuous Clearing Auction, or cca. Yeah. Talk about why you chose that.
- [55:24]
Zach Williamson
I could start on that. So we chose to partner with Uniswap on this because we thought it was the fairest way of doing an ico. As Joe said, what we really are focusing on is long term alignment. We want our community to stay with us and grow with us. And so the way we structured auction is very unique. We have a very long pre registration period to give people a lot of time to know that this is happening so that they don't have, we don't want to do this regular like what normally happens. Right. Three days, three year sale, lose a firm. Oh my God. Bye bye, bye bye bye. And what does that create? It creates massive information asymmetries between the buyer and the seller. It creates an artificially inflated price. Great. In the short term, absolutely terrible. In the long term. We want fair price discovery. And this is what the Uniswap's new auction does. It means that you know, if you, if you, if you get in early, yes, you do get a better price, but not so not such that like when you, if you get in late, you're going to get absolute, you're going to get a very bad price. Effectively your bid, if you bid early gets splits across multiple auctions that clear on different days across the scale sales schedule, which means that part of your bid will only occur on the final day. So the amount of the volume of bids will only ever go up on the days because you only have more people coming in. But the rate of change there is much more smoothed out with Uniswap's new mechanism. So not only is it fairer, but you get better price discovery too, which again we think is very important to make sure that people participating in the sale have. Have as much information as possible to make informed choices.
- [57:08]
Joe Andrews
I can add a few things, I think, like if we. We look at kind of other sales, a lot of like how you actually get tokens in the sale. It happens off chain. And so what you're really doing in like some of the most recent token sales is you're putting some ETH or some USDT or USDC somewhere and you're asking a centralized intermediary to give you an allocation in. In a token. And so that's not great for teams who've been building. We're meant to be doing stuff on chain and the reason we kind of wanted to partner with Uniswap is that they've been able to create a smart contract that does exactly that. It's kind of a smart contract where you can see exactly how you're going to allocate it. It's completely transparent, it's very fair and everyone's treated equally. So there's no sketchiness happening behind the scenes. It's. If you put ETH into this contract, you'll be allocated the same way as everyone else. And it kind of disincentivizes kind of these gas wars or these large whales kind of bidding and getting outsized allocations. And I think we've just seen that too much over the last kind of like seven years, both for airdrops and for token sales. And I think having a new approach is just refreshing. And this is where the industry needs to get to. It's a new product and so it's going to be exciting to see how it plans out. But yeah, there's actually kind of a paper that kind of proves it's like the fairest way to get on chain price discovery and the way it works at the end of the auction is we. Well, the auction automatically creates a Uniswap V4 pool at the final price. And so it's also kind of seeding on chain liquidity as a result and so driving kind of more users on chain. And so it's a really good way to kind of a find the fair market price, but also then create liquidity for the network. So if you're a sequence or approver and you have real costs, you need a source of liquidity. And so having kind of a pool which you can trade against ETH and Aztec tokens is very useful, especially if those costs are in eth. And so it's a good way to just kind of do things on chain and do things kind of, in my view, the right way, which is what the industry should be building towards rather than moving more things off chain. Yeah.
- [59:19]
Laura Shin
In terms of what you described about, like, the problems that we've seen with ICOs. I wrote about that in my book, like, for the 2017, 2018 era. So I'm very, very interested to see how this goes for you guys.
- [59:31]
Zach Williamson
Thank you. I think.
- [59:32]
Joe Andrews
I mean, I'd add a few things on ICOs because they'd come a long way. Right. Like before, like in 2017, 2018, like, we. We had some auction mechanisms, but they weren't kind of as sophisticated. And we also had no way to do kind of compliance checks. And so the reason we wanted to kind of do this on chain is now we have something like ZK Passport, and we also have a much more fair clearing mechanism. We can do this in a smart contract now. And so I think you get these rare moments where you can do things the easy way or you can do things the hard way and push the industry forward. And that's what we're trying to do here is just actually show there's a different way and it doesn't have to be like this kind of this whole historical meta. So gonna be very exciting in the next few weeks.
- [60:17]
Laura Shin
Yeah. I love that ZK Passport thing because it's like the industry always talks about how you can do that, and I feel like, oh, well, this is the first time. Or I could be wrong, but I think it's the first time that it's being done for a token sale. Right?
- [60:30]
Zach Williamson
I'm pretty sure it is.
- [60:31]
Joe Andrews
Yeah. In our knowledge. Yeah.
- [60:32]
Zach Williamson
I mean, and there's. There's a reason for that. You know, we. We again, we talk about this, like, you know, on chain, permissionless id, ZK stuff. So we're like, well, we better. We better practice what we preach. Right. We better use our own products, or at least ecosystem products.
- [60:44]
Laura Shin
Yeah, yeah. Because, I mean, that example of, you know, oh, like, you can go to a bar without showing your address or, you know, your actual birthday or whatever and just prove that you're eligible to walk into this bar. Like, I think people were saying that to me in 2015, you know, and so, yeah, it's just cool that. That you guys are actually using it and out of curiosity for that reason, like, because a lot of the token sales, like, in the US It'll be like, only accredited investors or they don't allow us people to participate. But, like, are you. What are the restrictions or what are the requirements?
- [61:18]
Joe Andrews
So the token sales open to pretty much all Countries like US Retail can participate. And the main reason that's the case is because there's an actual use for the token. So the use for the token, if you buy it, is you stake it and use it in the network, or you use it in governance or for fees. And so because we've, like, been building this for seven years, like, the token is the last piece of the puzzle. Puzzle. And there's, like, a use case for it. So because of that, we're able to open up to nearly all countries in the world, except sanctioned countries, which is exciting.
- [61:49]
Laura Shin
Okay, cool. Yeah. Because I wrote about how there was, like, this period where it would be like, you know, the countries that are prohibited from participating are, like, North Korea, you know, Iran, like, whatever. Whatever the countries were, and the United States, you know, and it's like, what? So anyway, that's. That's awesome.
- [62:08]
Joe Andrews
Yeah.
- [62:09]
Laura Shin
All right, so I actually want to talk a little bit about the language. So you've incubated this privacy programming language called Noir, and the tagline is that it's like Rust for zk.
- [62:22]
Zach Williamson
Yes.
- [62:23]
Laura Shin
So explain, like, why ZK technology needs its own programming language.
- [62:27]
Zach Williamson
Yeah, for sure. Because it's been a journey, right. It's been five years of building, and it would have been lovely if we didn't have to do it. But the challenge is not for us. It's not just zk, but it's also private, persistent state. What we found very early on, with our own experience writing ZK writing circuits, is that there's a lot. The semantics around how to, like, write programs around private information are very, very different to regular programming environments where you need different keywords, different, like, different kinds of, like, semantic flows. You need different programming patterns. And that's. So we looked at a language like Rust and we're thinking, can we just extend it? Can we just make this a Rust library that people can write in? And what we realized as well, so we'd have to extend the Rust language so much, add our own stuff in so much that it wouldn't be Rust anymore, so we might as well do it from the ground up. And the analogy I like to use is solidity. Ethereum chose its own language because it had all this custom stuff it needed to do, it needed to handle. You need to know who your message sender is, who the block hash is. You need this concept of a contract, and you have all of that with Aztec. But with privacy in general, you also have similar things to do with private stake. You know, if you. If you want to. If you want to handle private information. You need like non inclusion checks in Merkle trees. You need. Everything is encrypted. You need to perform like weird, like encryptive primitives, hash function algorithms. You need a lot of like macro piece meta programming. Another thing is, you know, you need. You. You likely want to split your program into kind of two sections. One where you're. You're performing checks on the results of a computation which does need to be zk, and then you're actually performing the computation itself. And maybe that actually doesn't need to be proven. Like there's parts of your program you can not prove and just check the results of. Which means you need, well, different keywords. For example, we have utility functions in Noir which perform this purpose. And so we just basically thought we want control over the semantics of this language so that we can create an abstraction layer that's as easy as possible to use. And we felt we couldn't get that by extending an existing language and it was better to do it all from scratch.
- [64:47]
Laura Shin
Okay. And I saw. So I was a little bit confused because I think I saw a tweet saying that it was going to reach its 1.0 or you were going to release the 1.0 in. In Q1 of 2026.
- [64:58]
Zach Williamson
Yes.
- [64:58]
Laura Shin
So that makes it sound like that's the first version, but you have already built this chain so well.
- [65:06]
Zach Williamson
I mean, there's this version 0.01, you know, 0.1, 0.2 up to 0.2607. So we've just been a little bit conservative about what we're calling version one, because version one is audited and therefore we can say, yeah, it's audited, everyone can use this. Whereas right now we need to put some health warnings on going, yeah, you can use it, but it's not audited. So if it breaks, don't. Don't come at us. We've told you it's. But so, yeah, version one is it'll be audited and that's when we're competent in it.
- [65:33]
Joe Andrews
We've had Aegis sets so like the kind of like 1.0 release candidates for I think four months now. So, like all the features are there. We're now, I think for the first time putting a new language through audit. And so we're being cautious on the timeline. The audit's underway already, so it's kind of just how long that's going to take, which is where it's going to get to, I think just to add to some of Zach's points earlier, like, we ran into this problem building Aztec itself, like our speed of iteration and development. Like, Aztec used to be written in C. Like all of our circuits were in C. And we had some tooling that Zach wrote and the team wrote and like, it was like, very hard for people to like, move quickly because it was like it was custom tooling for, I guess, a language that wasn't fully fit for purpose. And midway through, like the Noir journey, we switched to writing Aztec itself in Noir because Noir is a generic language for any ZK application. And the speed of iteration just went up. Like there was like, probably we doubled the amount of people within the company who could write Noir circuits. And I think there was an original estimate of like how long it would take to rewrite what we'd done in C in Noir. And I think the team did it in like 10 days or something. Crazy. Yeah. So it was amazing to see, like this actually come. Come into fruition. So I think the tooling side of it is the other side. It's not just like the features of the language. The tooling you need when you're dealing with like encrypted state and private state is very different to kind of like a normal language. So you need a whole ecosystem that develops around the language for it to be successful. And so it's one thing having kind of the features, but then it's also the tooling, the libraries and yeah, the community. And so we had our third noircon at devcon and so we have this whole amazing ecosystem of people who, they're building things with Noir. They could be for just off chain applications. You can make proofs for anywhere people deploying on Ethereum. There's applications on Ethereum that use use Noir and then Aztec obviously is like the natural home for deploying those applications because it's, it's designed for it and it's just amazing to see kind of it exploding, I think as people start going down the Noir rabbit hole. Yeah, yeah.
- [67:42]
Laura Shin
I like listening to you talk. It just reminds me of even learning about how Vitalik came up with Ethereum. You know, it was like he, his, his unlock was realizing, oh, I need to write a programming language. So it's just. Yeah, it just feels so similar. So does this mean that, like, all of the apps built on Aztec need to use Noir or, like, can some of them use solidity or like, so.
- [68:07]
Zach Williamson
They all need to use Noir because everything needs to compile down into ZK proofs And right now Noir is the only way of doing that. You Know, in theory, one could write another. Another language. So because we had this separation of concerns with Noir because we wanted it to be a public good that anyone could use, so we didn't want Noir itself to be too closely tied with astec. Noir has a separation where we call it like a language front end. So it doesn't, nor itself doesn't produce zero knowledge circuits. It produces something we call a CIR Abstract Circuit Intermediate representation and is very similar to LLVM for computers, where the idea is you can take a cryptography backend and turn the ASEA into ZK Circus, ZK proofs. So we do that with our software we call Beresenberg. And so in theory, somebody could write another language that compiles down into a sir, and you could then use that to create Aztec Smart contracts. And I'd love it if. It'd be amazing if somebody took on that challenge because, like, diversity of options is always a very good thing. Um, but right now, it's only Noir that can do that.
- [69:16]
Laura Shin
Okay, so I absolutely have to ask you about some regulatory stuff because I was, you know, as you were building this, I'm sure you were watching all this stuff happening with Tornado Cash.
- [69:27]
Joe Andrews
Yeah.
- [69:28]
Laura Shin
So what was that like? You know, what were you thinking? Were you worried about what you were building? Did. Was there a certain point where you finally felt comfortable releasing this or like, yeah. What was that like along the way?
- [69:40]
Joe Andrews
Yeah, maybe I can take it to start with. I think it definitely. So we had, like, products live during that whole period. We had a previous product called Aztec Connect, and it definitely was like, no, it was kind of jarring to see something that people took as like a normal thing being challenged by a government. I think for us, it didn't cause any kind of immediate issues other than a decision we had to make, which was, should we, like, could we keep running this thing as we were, or did we have to decentralize it? And so it kind of maybe accelerated our timeline to, I guess, like, moving on from that iteration of Aztec to building this fully decentralized programmable version. I do think that as an industry, as a whole, it put the brakes on a little bit of people's appetite to develop with privacy. We've been very convicted for seven years, so we were able to kind of convince the team and just build through this. What we're building is legal. It makes sense. It actually lets you build more compliant applications. But we did see as a whole, a lot of developers outside of Aztec kind of maybe shy away from the ecosystem. But I Think in terms of compliance in general. The whole thinking behind building a private execution environment is that it can actually be more compliant if you can tie your identity to a cryptocurrency account. You can actually kind of build applications where sanctioned people can't take part, because you know who everyone is in a private world. And so I think we unfortunately existed for a few years in this period where privacy was kind of being attacked. But I hope in the next kind of like three years, we'll see applications that can take advantage of the expressiveness of a chain like Aztec to build compliant applications if they need it and still uphold strong privacy guarantees.
- [71:32]
Zach Williamson
Yeah, I can share my experience as well. I mean, I remember this all going down.
- [71:35]
Joe Andrews
Right.
- [71:35]
Zach Williamson
And the arrests happening. And I think my initial reaction was this anger, you know, because, you know, I know these people.
- [71:41]
Joe Andrews
Yeah.
- [71:41]
Zach Williamson
And, you know, when you see public figures on, like, TV saying that these people are terrorists or they're helping terrorists, I'm like, no, they're not. So maybe very angry, maybe want to double down on what we're doing. And, you know, I think obviously it was a challenging time to develop, and as Joe said, but even if Genza was director of sec, we were still launched. We just probably wouldn't be. The token sale wouldn't be. Probably not open to us people. And we, Joe and I would probably stay away from America, but we would still launch because we are permissionless, decentralized network. You know, there is no valid, legitimate ground to have a go at what we're doing. And so we're willing to fight that battle. But I also think, you know, at the time, it didn't slow us down, not really, because I guess this is where online narratives and reality collide a little bit. We're not Americans. We're based in the uk and despite. And, you know, you might think, oh, my God, that's the. That's the worst place to be. And it's like, no, no, it's been fine for us. It's been absolutely fine. You know, and similarly, you know, like, man, you know, this industry is going to have a lot of historical trauma about the Genesis era. Because, for example, if you look at the spectrum of noir developers and people writing zk, a lot more Europeans than Americans, because the Americans for a long time have been scared. So, yeah, that's. Yeah, the SEC was not doing its part for American dynamism, that's for goddamn sure. But it just highlights how important this stuff is to execute on in build and ship. Because, you know, if certain powerful incumbents believe this technology to be a threat. That means we're onto a good thing. You know, I don't believe for a second that the attacks against these individuals were because of North Korea. Like North Korea was the justification, it wasn't the motivation, I think.
- [73:35]
Laura Shin
Oh, interesting. All right, well, last question and I don't even know if it's too early to ask this question, but are there any specific apps that are being built on Aztec that you're already excited about or, and if, if it's too early to ask then just, just tell us whatever it is that you're excited about for the next, I don't know, six months or you know, new milestones.
- [73:56]
Joe Andrews
I got some apps that I'm excited about. I think the obvious one is kind of this idea of like private intent, so bridging, bridging kind of assets privacy to all of Ethereum. I think people don't quite understand. Like you know, if you ask someone what percentage of transactions on Base or Arbitrum or Hyper Liquid would like to have privacy, I think the answer is pretty high. And like we now have a system that will be able to kind of like deal with those transactions and give users a form of privacy that they've been craving in terms of like Aztec specific applications. So like things that will be deployed on Aztec. For a long time we've had over characterized lending as like the only source of lending on blockchains. There's been some, some products trying to do RWAs, but there's nothing for kind of like retail consumers like and if you look at your kind of like traditional Web2 alternatives, you can get a loan that's secured by who you are, your salary, your bank account and we don't have that on crypto today because that's very sensitive information. Like I'm not going to put my bank statement into an Ethereum transaction so everyone can see it. There's loads of people kind of exploring this idea of like consumer finance on Aztec and in Europe we have this standard called open banking and all of the bank data is signed by the bank institution. And so you can basically feed your bank statement into a zero knowledge proof and start to prove things about it. You can say my salary is above X, my outgoings are below Y, so I'm credit worthy and you can give me a loan. And so I think we're going to see like an explosion of I guess real world finance being built that I find very exciting. And then like the, the other basic ones are just payments in general. Like you had a Podcast I Think recently about X402. I think X402 without privacy is incredibly dystopian and so like being able to.
- [75:46]
Zach Williamson
Not gonna get used very much.
- [75:47]
Joe Andrews
Yeah, just being able to see every website I pay for things on like is in the real world as, as well. Like whether it moves kind of out of kind of like just websites and into like shops and commerces is fairly terrifying if you can see like the, the balances associated, associated with that. So I think actual private payments is really exciting.
- [76:07]
Laura Shin
I don't know if you remember this, I mean this was a long time ago, like maybe 2014 or something, but there was this New York Times magazine article about how there was some teenager who was pregnant and Target figured out that she was pregnant before her dad knew and sent a, like a, a coupon and you're like oh for your baby or whatever and the dad was so mad. But then it turned out that it was right and, and so yeah, just what you're saying, like I'm like oh my God, it's like that.
- [76:36]
Joe Andrews
Yeah, yeah, it's going to be, it's going to be crazy. And like especially in like in a world where we now have like AI models that like your browser is an AI model with like these new kind of browsers, all of that data is being hoovered up without people even knowing. It's. Yeah. I think we need to kind of fight back a little bit against a lot. Basic privacy standards.
- [76:55]
Zach Williamson
Yeah.
- [76:57]
Laura Shin
And your apps that you're excited about.
- [77:00]
Zach Williamson
Yeah. So I think, I mean we lost 16 teams building on ASIC right. For testnet for launch, so. Oh man. If you'd asked me at the start of the conference, my brain wouldn't be so addled. I mean like the base, honestly I'm excited for just the most basic stuff because you start at the grand layout with basic cryptocurrency transfers, grace of payments and so like the Obsidian and the Asgard wallet teams building on Aztec, they're really pioneering. They're doing trailblazing stuff, they're doing account abstraction where you author, you login with a passkey and very other novel ways of authenticating yourself and building the seeds of a deep bank where you can do your basic trades but also eventually you want to get that connecting to lending protocols and eventually might take a bit more time get this plugged into real world assets. So it's not just synthetic instruments. I think private stablecoins, private stablecoin issuance as well and on and off boarding as well as private transfers is very Very important because it leaks so much information. If you're using stablecoins to settle like expensive trades, for example, it's all tied up with having. Needing private portfolio management. And you know, this can all be done on Aztec. Might take a little bit of time to execute on but you know, I'm very excited by the prospects and something we're determined to. So it makes sure.
- [78:15]
Joe Andrews
What about games? You've been tinkering with some games and. Oh yeah, like CK interest.
- [78:19]
Zach Williamson
Yeah, I'm too boomer to make a bit like to make actual compelling games. No, but there's a big problem the games that everyone else likes. But ZK gaming on chain I think is going to be a very important growth sector because like games without information asymmetry are boring. And if you want to do things that are really on chain, then you need privacy for that. Information asymmetry. Even something as basic as you remember, cryptokitties, you know, this is the first NFT that really came up and you can't do cryptokitties today on a transparent blockchain because they relied on code obfuscation to create certain like types of combinations of cats that were rare or valuable. But now that the stuff gets reversed engineered instantly because it's just, yeah, it's easy to figure that stuff out now where the sector's more advanced. But you can bring that back with CK and privacy and things like Tetseo, which we're partnering with to get like, trust us, like random number generation in a secure way. Which means you can genuinely create NFTs where you have information asymmetry, where you can get an outcome on chain that you. Where you don't know what that outcome's going to be when you send the transaction. And I think that's going to be quite a powerful primitive.
- [79:26]
Laura Shin
All right, well, this has been so fun discussing this all with you and congrats again on launching the chain.
- [79:31]
Zach Williamson
Thank you so much.
- [79:32]
Joe Andrews
It's such a pleasure. Thank you so much, Hannah.
- [79:34]
Laura Shin
Good luck everyone.
- [79:36]
Unchained Promo/Ad Host
Unchained is produced by Laura Shin, with help from Matt Pilchard, Juan Aranovich, Margaret Curia and Pam Majumdar.
- [79:43]
Joe Andrews
Thanks for listening.