Gnosis catch-up with Stefan George

Welcome to Zero Knowledge. I'm your host, Anna Rose. In this podcast, we'll be exploring the latest in zero knowledge research and the decentralized web, as well as new paradigms that promise to change the way we interact and transact online.

This week I chat with Stefan George from Gnosis. He walks us through what the Gnosis team has been building over the last two years, and we cover some of the projects that have since spun out of the organization. We then chat about the evolution of the Gnosis Chain, the work they're doing to make bridging more secure. And finally, we talk about Gnosis Pay their new product focused on the original blockchain use case that of payments. Through our conversation, we get a chance to cover a number of relevant themes, such as intense account abstraction, ZK bridges, decentralization, and we make a few predictions about what the larger Ethereum ecosystem might look like in the future. Now, before we kick off, I just want to let you know that the application to attend zk10 is now open. This event happens in London on September 20th. As always, we aim to bring together the top researchers and engineers working in ZK to share their latest research and new findings. Check the show notes for the application form and please note spots are limited and only folks who applied will be offered a chance to buy a ticket. So do get your applications in soon and yeah, hope to see you there. Now Tanya will share a little bit about this week's sponsors.

Polygon Labs is thrilled to announce Polygon 2.0, the value layer for the internet. Polygon 2.0 makes mass adoption possible by offering users and developers unlimited scalability and unified liquidity. This mission is fueled by groundbreaking ZK innovations, including a first of its kind ZK powered interoperability protocol. And the next generation of the industry leading and widely adopted Plonky2 proving system. Polygon 2.0 will change the way we experience Web3 by bringing the security and decentralization of Ethereum to the scale and usability of the internet itself. Polygon 2.0 and all of their ZK tech is open source and community driven. Reach out to the Polygon community on Discord at discord.gg/0xPolygon to learn more, contribute or to join in on building the future of Web3 together with Polygon.

Aleo is a new Layer 1 blockchain that achieves the programmability of Ethereum, the privacy of Zcash, and the scalability of a rollup. If you're interested in building private applications, then check out Aleo's programming language called Leo. Leo enables non cryptographers to harness the power of ZKPs to deploy decentralized exchanges, hidden information games, regulated stablecoins, identity products, and more. Aleo's incentivized testnet is now live. Participate as a developer, apply for a grant or go for a bug bounty. Check out aleo.org/blog for more info. That's aleo.org/blog. You can also find the link in our show notes. So thanks again, Aleo. And now here's our episode.

So today I'm here with Stefan, who's one of the three co-founders of Gnosis. Welcome to the show, Stefan.

Thanks for having me, Anna.

So I've already had your other co-founders on the show over the years, but I'm very excited to get a chance to chat with you. Just to give folks a sense though, like the first time I spoke to one of your co-founders, it was with Friederike and we were talking about the original idea for Gnosis, that is like as prediction market. I then chatted with Martin about CowSwap and the DAO work you were doing. I had a chat with the Zodiac project, which I think spun out from Gnosis.

Not yet, but about too. Yeah.

Okay. And I also had a chance to chat with Martin again last year, just after the, like we talked about the fallout of Tornado. So with members of your team, I feel like we've covered a lot of ground.

Yeah.

But today, I mean, what I'm hoping to do is do obviously a recap since the last time we checked in on Gnosis, but also find out, you know, what are the different building blocks? What's spun out, what's still inside, what are you guys working on? The thing is about Gnosis is that I feel like you've consistently and kind of quietly over the years shipped some pretty interesting, pretty important pieces of software. I think there's a lot of people who are interfacing with Gnosis software, but they might not know that. Yeah. And this has mostly been in the Ethereum community,

Right? Yes. Yeah, I know. Thanks for the intro. Yeah, I think for Gnosis the interesting story is that we continuously pivoted or we did just many different things in parallel. And I think the reason for this is that we have many different interests, at the same time I think we are first of all entrepreneurs, so we identify problems and we try to find solutions for it. And you can imagine at the beginning of Ethereum, which was also the beginning of Gnosis in some way, there were lots of problems. There were only problems to be solved. And so we picked up a few of them that were for us interesting. And that we also found instrumental to actually accomplish the core mission of Gnosis, which was at the beginning, prediction markets. So for prediction markets to work, you have to have wallet infrastructure.

That's why I started the Safe project within Gnosis. And then of course, you need exchanges in order to trade prediction markets. So we started very early on researching and building decentralized exchange infrastructure for Ethereum. And so it was just not necessary to build those in order to actually accomplish what we actually want to do at the time, which was prediction markets. It turned out that less people actually care about prediction markets. People like to theorize about them, like to theorize a lot, but then when it's actually finally about launch, then actually very few people cared about it. And there are many reasons for that. Fortunately, however, everything else that we built in order to make prediction markets work worked actually much better than prediction markets.

So it's almost like the tooling you built to get to your goal became your business.

Exactly. That's exactly what happened. And eventually those kind of side projects became the main projects and at some point it became so big that it became obvious it would be better off growing faster and being more successful by not being part of the core organization anymore, but being independent creating their own identity and grow as they think it would be best.

That helps us to understand a little bit why, you know, even in this early intro, this idea of spinning things out. So I have a sort of a starting question, which is, do you now consider Gnosis to be a bit of a like company builder or almost like a studio? Because you have these ideas, you build the product, you build up a team, and then you spin them out. And I know this isn't like the way you started off but do you feel like you could kind of call what you do now? Something like that?

Was a venture builder

Was a venture builder. Yeah. And now many people refer to us as actually being more of what Consensus wanted to be. So I think it's just kind of in our nature of being entrepreneurs having many interests and yeah, trying to align many people with this vision. And then eventually those become too big for us to continuously drive all of them forward. And we find the right people that can take on ownership and become founders of those. I think that's the extremely important part that many people completely underestimate. Of course, you can always easily say like, okay, this becomes now spinoff, but you need people that can actually drive this. Like, you need founder type people. Like we fortunately had Gnosis and are having Gnosis that are able to really drive things forward that form their own vision, that are able to hire great other people to work with them. And if you don't have this then, well, there's no spinoff.

I wonder, can you share with me some of the projects that you did start that didn't go anywhere? Because I feel like there's no way it's only home runs. We talked about some like big ones.

It's a very big graveyard, usually we clutter over this, but no, I'm happy to talk about it

Can I ask you about one of them? And maybe it wasn't a graveyard one, but I do remember many, many years ago we were talking about like auctions.

Yes. You're talking about the Dutch Exchange.

Dutch Exchange did that, was that a graveyard project?

Yes.

Okay.

That I can fairly confidently say it was a graveyard project and I think it was probably, if you compare the like time we invest into it to where it is now, like the biggest investment of resources from our side with the smallest like output or success.

It was that one.

It was still instrumental for CowSwap. So in some sense you can see many things as being continuation of other previous projects because it's the same team working on those. They have the same, like they gained experience through those failures, obviously. And ultimately led to actually better products. And I think also what we saw is like, we hired a lot of people, which were great engineers, but they were new to blockchain. And so they mostly relied on us giving guidance and making design decisions. And I think over time of course, it became more and more senior so they could really much better argue with us which ultimately led also to better products. And I think for the Dutch Exchange specifically the, yeah, there were many, I would say user experience problems, why it never really took off. And it was also an ownership problem. So as part of Dutch Exchange, actually, we also formed DXdao, if you remember

DXdao

So DXdao

DXdao

DX. Like that's why Dutch Exchange DAO.

Got it, got it, got it.

And so the problem that we foresaw with decentralized exchange is not only writing the smart contract and writing an nice UI, but you also need someone to host this UI and be responsible for the operations of this exchange, right? And we didn't have any license for this. So together with launching the exchange, we also launched a DAO.

Okay.

And we had one of the first liquidity mining programs actually that gave those people ownership in the DAO.

Wow.

And this is how the DXdao was formed. So it's a very interesting experiment. You can think about this like you are in a room that's dark, like this room without light, you switch on the light, and then you see all the other people in this room. And those are your co-founders.

That's kind of how the DXdao came into existence. So suddenly there was a DAO, there were people with ownership, but no one knew each other until the point where the liquidity mining program was actually done. And you could really form teams. Well, unfortunately, this as with every DAO the struggle in getting started. And so they were not really able to drive the exchange to being successful. So Uniswap was way more successful, obviously, but DXdao itself had interesting activities. So they really started forming an organization, and most importantly, they were even able to coordinate a token sale.

Okay.

And

For themselves, or for someone?

Well, for the DAO itself.

Okay.

They capitalized the DAO with a significant amount of Ether, I think something like 20,000 Ether. And yeah, they gave ownership in the XD tokens. And yeah, it was very interesting to see for a couple of years. Unfortunately now it ceased to exist because ultimately they were not able to drive this organization to be successful in terms of developing interesting successful products.

Ah, okay.

So actually just a couple of months ago, there was a redemption process where DXD token holders could redeem their tokens for Ether. And yeah. In some sense I would say DXdao has failed and it's now also ceased to exist. But it's definitely one of the most interesting experiments that we did.

Did this lead at all to the Zodiac stuff? Like, did you, all of sudden?

In some sense you can say yes because some people from the Zodiac team obviously worked on the DAO related tooling that was used for the DXdao to come into existence but more importantly, the CowSwap team worked previously on Dutch Exchange. And then a lot of the learnings about how the auctions were run and also especially on the user experience side, led to the creation of CowSwap.

Okay.

Which then obviously became successful enough to actually become a spinoff.

CowSwap. It sounds like there's a lot of pieces that sort of like you needed the exchange for the prediction market. That was sort of one driver. The other one is you had created this Dutch Exchange, but it wasn't really the project that people stuck to. So did they switch to CowSwap or did they really like graduate to it?

So it was more like a switch, I would say, or like if they use Dutch Exchange at all. So the problem with Dutch Exchange was so the way our Dutch auction works, you start a very high price and then the price goes down over time until a time when everyone's bidding and the auction's finally closed. So from user experience point of view is kind of annoying. If it was want to buy a token, I don't want to wait like 6 hours or so to finally have the price point at the mark where I want to invest into. So that was just not really great. And for us it was mostly a mechanism to find a fair price, but there are better mechanisms.

Okay. But did CowSwap use that? Does CowSwap like under the hood, does CowSwap have that?

So no, CowSwap is using different type of auction under the hood. It's like a multi token batch auction designs actually much, much more complex. But the team in that we hired for Dutch X was, I would still say kind of junior in a sense in terms of experience in blockchain. So something as complicated as CowSwap I think could only be formed by a team that already had a lot more experience. And I think Dutch X gave the entire team enough experience to even work on something as complicated as CowSwap.

I know that what we want to talk about today is more about like Gnosis Chain and some of the upcoming stuff, but I do, I want to sort of stay on CowSwap for a second because just recently we had Chris Goes on the show talking about Intents. I always feel silly saying it because I think I sound like I'm saying Intents, but yeah, when I started to learn about it, the example that was highlighted for me, if you want to see an intents based system in the wild try CowSwap, which I use actually quite often. So I already knew it and I was like, oh, that's what you mean. Like, you kind of put out what you want and you kind of put out how much you want to pay, and then it finds it through some much more elaborate, you know, going through different tokens to get it.

That's exactly what CowSwap is. So ultimately if you trade on CowSwap, what you sign cryptographically is you intend to swap one token for another token for a certain price and a certain price range. And you can also define like a timeframe which you want to do this. And then this order or intent is out in the wild and what you call solvers in CowSwap they can take those intents and try to match them with other intents to finally settle them on-chain. So in that sense, it's actually an intent if you would trade on Uniswap, for example you're actually signing a concrete order, Ethereum order or transaction, which directly defines the path of execution on-chain. So that's very, very different. Here we kind of are abstracting a way of like the concrete way to settle to just define what's the user intent. And that of course allows much more flexibility in terms of how this intent is actually materializing in a transaction.

And, but do you think, like, have you been following the intent conversation then? Like how do they take what you've done and push it forward?

So this is like a very complex topic. I think it's also kind of related to the general idea of also even account abstraction or like how you, I would say the more complex the system becomes the better it is to ask user for their intent instead of the concrete execution.

Okay.

Because there can be many factors playing into this and if you just have the intent, then you have the most abstract form of what a user actually wants to accomplish in the system. And that's only what the user actually cares about.

Totally.

And it allows us to build the underlying systems to fulfill this intent. And they can become arbitrary complex, but we don't lock the user into like one concrete execution path. That's kind of high level how I would explain it. And that's why I think also it's a better way to execute.

A very simple example of this why I related to account abstraction. One fundamental feature that account abstraction also through safe contracts allows to do is to abstract the intent from paying the fee for the transaction. So one big user experience issue on blockchain is still people have to understand what's a fee for transaction? How do I price this transaction? What if there's a gas spike I have to resubmit this transaction because my gas price was too low. You know, all of those user experience issues, it can all be now obstructed away because the user only cares about swapping a token. They don't care about the transaction fees and the account abstraction allows exactly to do this. You just have your intent. And then someone else, like a relayer, searcher, solver, you name it, many terms for this can pick up this intent, turn it into something that actually executable and make sure it is included and the user doesn't have to worry about it. And that's why intents are so important.

I want to talk a little bit more about account abstraction.

Yeah, for sure.

A while ago I had Argent on the show and we did talk about sort of the smart contract based wallet, but isn't Safe already kind of account abstraction. Or is there, like what is, why is there then a conversation about it now? Are people trying to get more people to move to these types of wallets or I'm just curious your thoughts on that.

And it can't have the name account abstraction anymore.

No

It needs to be so hidden.

Yes, it is. So actually I would say the Safe is most likely to be the core infrastructure for this, simply because it's a most widely known smart contract. So for example, every Worldcoin user is using a Safe contract and can, well different opinions on Worldcoin but I'm very glad with the design decision to use Safe contracts. And now we see more and more wallets implementing Safe contracts under the hood to allow account abstraction and no Argent's also a very good example. Of course, they have their own system and they are mostly, I think, focused these days on zkSync and Starknet whereas we are kind of agnostic. So Safe contracts exist on every, on every EVM network at this point.

Yeah, I bet.

And there are lot of mobile wallets now starting to implement Safe contracts because you understand it brings a lot of advantages for account recovery, for fee abstraction and all of those things that will be necessary to onboard more users.

Right. Got it. It's like a standard,

The MEV stuff

Exactly which is because it makes things more efficient. And of course intents also include MEV, right?

Yeah. Wait, what do you mean? Like it adds more MEV it creates MEV.

There's a lot in what you just shared. So first of all, this idea, so I'm realizing and talking to you that I have not visited Ethereum land in a while because I mean I've talked to maybe individual L2's or like, but I feel that the idea that an EIP does not succeed on the main but only succeeds on the L2's, this is sort of the first time I'm realizing that that's possible. That really talks about like a, like a difference in, I don't know if it's like communities or like builders will have different choices to make if they are building on an L2 in this case. And that's something, I mean I feel a lot of the L2s are sort of selling themselves as like one-to-one as a virtual machine, one-to-one. But there is different activity actually happening there.

Absolutely.

Interesting.

So I think it's very obvious that Ethereum will not be a place for individuals in the future to operate. It will more for institutions and Layer 2's

Wow.

As like a kind of settlement layer and a place to rebalance between different parts of the network but it will not be you or me making a transaction. It will just be too expensive.

And then you also talked about this like centralization of mempools on the different L2's. This to me is probably like a whole episode to talk about like the role of the searcher and what that means going forward. I wonder if now would be an interesting time to also introduce Gnosis Chain though beause I know we mentioned it, right? But I feel like yeah, because that's what you are currently working on in a lot of ways. This is where the project has evolved to. I do think we did mention it on the episode with Martin the last time he came on. Right. But yeah, let's, let's introduce Gnosis Chain.

Very, very briefly. So Gnosis Chain previously was called xDai. So many of you probably know it from the times when you were issuing POAPs

Yes. Also, we did an episode on that a long time ago.

Great

So if people want to hear that exactly, just I think that was the first time I understood bridges actually.

Okay, great.

Because that was very much a bridge. It was like another network. That was bridged and that was the terminology I think they were using.

Exactly. Like basically Gnosis Chain is the, let's say original side chain to Ethereum.

Okay.

It was the first one before Polygon, Binance Smart Chain. Any of those existed.

And it was called xDai.

It was called xDai.

And it was associated with this burner wallet phenomenon.

That's right. Burner wallet was one of the main applications really nice to use for payments. And so it always kind of was associated to the use case. We saw it being an underappreciated ecosystem and we wanted to help xDai to succeed. And so eventually Gnosis and xDai merged, it was renamed to Gnosis Chain and we changed the technical roadmap. Previously we was using POA Consensus and we changed it to actually Ethereum Beacon-chain Consensus. So we have our own Beacon-chain running for Gnosis Chain. And what makes the Ethereum consensus very interesting is that it allows for a very decentralized validator set. So Ethereum has about, I think 550,000 validators right now. Gnosis Chain has about 130,000 validators. And there's literally not any other networks that come close to those numbers. And of course this number also has to be taken with a grain of salt because it's more important how many different entities actually running those.

By other networks, do you mean L2's or do you mean the, like other L1's?

Actually all of them. So Layer 2 is actually very centralized. There's only one one sequencer.

Usually, although there, there's talk of decentralizing the sequencers. Every single one has that on the roadmap.

Of course, of course. But then you kind of go back to the same problem. Right. And so it's just like kind of having a blockchain built on a blockchain. So of course everyone wants to decentralized but de facto just looking at what is the state of networks today. Like Layer 2's are very centralized. And then you have Polygon, which of course has a lot more validators, but also unclear who's actually running them. How decentralized is actually then here, VC chains like Solana, it's very obvious. Very few parties can conclude to actually compromise the network. So decentralization is actually a very scarce resource. And only Bitcoin and Ethereum fulfill the criteria, I would say, to have a really fully decentralized network. And at the same time we see Ethereum block space is in high demand. It is one of the few networks where people have a lot of trust in this network.

They store their net worth on the network they are willing to operate globally. And so we saw an opportunity to create more of this decentralized block space by offering actually highly decentralized network following one-to-one the Ethereum roadmap. So we are the only, like one of the very few networks that is actually running the same stack as Ethereum, which also allows us to benefit from all the research that is done on Ethereum. So for example very soon we'll have Proto-Danksharding being implemented. And that requires actually Beacon-chain. So we have lots of networks that are EVM compatible, like buying the smart chain Polygon, but they can never implement this EIP because they don't have a Beacon-chain.

You say you're one-to-one, you're sort of following the roadmap, but are you sometimes able to do things before the Ethereum network can do it? Like that was, for example, did you go to proof of stake first?

And on this layer, I think we can experiment a lot. So we experiment with designs that allow for example, encrypted mempools to allow more privacy and those kinds of designs, we can innovate on top of Ethereum without being directly incorporated into the core layer. If it proves to be successful, eventually we get enough mind share to drive this back to Ethereum and say like, here, look at this. This is something that makes sense, it got adoption and then eventually can be implemented in the core layer. But that's a very long journey. So I would dare to say we are going to front run Ethereum. We are going to be modest and follow and make sure that we are actually on top of things.

Did Gnosis, like another idea, and I don't know in which interview when I heard this, but wasn't Gnosis Chain kind of early on also meant to be like the place for more non-economic activity, sort of like the example that was often given was like voting on a DAO, right? You would do some sort of the activity on the L1, but like the voting and that all of those signatures and all of that activity would happen on Gnosis Chain. Are you still thinking about it that way? It sounds like not really.

So one way we see Gnosis Chain is to be sort of an extended version of Ethereum. So as we are very decentralized and we have a very secure connection to Ethereum using actually ZK technology. We can be used as a space for trustworthy computation that can eventually be relayed to Ethereum. So that's something we still are very excited about, and we are also working with those teams that try to enable us including like Axiom for example but those

I was about to say, as you said, that sort of the extension, it sounds a little bit like the co-processor or the

Exactly.

But what they're doing is so specific and it sounds like Gnosis Chain is way more generalized

Yeah, Gnosis Chain is very generalized, but Axiom I think is a very great team that can enable some of those use cases. And so we have working closely with them to enable them also specifically between Gnosis Chain and Ethereum. We are also talking actually to some Layer 2's to use Gnosis Chain as a place for data availability.

Whoa. Oh my gosh. That's such a flip

As Ethereum. Oh no. Just, it's kind of not a flip

But I think of it like, isn't DA always meant to be like, oh, I mean, at least the Celestia example is like data availability is sort of the center and then you have rollups that come off that and all the execution lives on the rollups

Well, that's Celestia's vision.

Okay, tell me yours. Wow.

Like I mean Celestia yeah, great project. But of course right now Ethereum rollups like using Ethereum for data availability and now we are talking Layer 2's that are considering using Gnosis Chain for data availability instead.

But you would be using data like in this case. Are they then writing data to Gnosis?

Yes. Yeah.

But isn't the reason you do it on Ethereum is you want the like, amount of security that Ethereum has?

Well, that is the pitch that Vitalik is doing. Of course it comes at a very high cost because ultimately your costs are directly proportional to the general fee cost on Ethereum. So potentially they're very high, like very costly to operate, and from many use cases I think will not be viable. And actually data availability is in our view, requires much less security assumptions compared to computational integrity. Right. Like you want to make sure that you don't temper transactions by settling on Ethereum and should be able to challenge this on Ethereum. But where the data comes from and that you have data available for this, you can use many different systems. It doesn't have to be Ethereum, it can be Celestia if once it's available it can be Gnosis Chain. So it can be any of those systems. And important is that they're security connected to Ethereum to be able to prove that something was available. And that's something that we are also developing, obviously, like a secure bridge that allows to communicate between Ethereum and Gnosis Chain.

As you're saying all this, would EigenLayer also kind of be in that competitive sphere that you're talking about? Or is it closer to you than Celestia in a way?

Good question. So I think it's a bit closer to us for sure but it plays in the same, in the same field.

Wow. I feel like I need to see a map

Me too.

I mean, even when you, when you talk about data availability of L2's, like if they're using Gnosis and Gnosis meanwhile is using the Succinct technology to Ethereum and they're using their own sort of like ZK, I mean, not really bridges, but you know, the zk-rollup proof, right? It's like you see sort of the central Ethereum, you see all these chains branching off it, and then you see all these things between them. And I always thought, oh, it'd just be bridges between them for tokens. But now you're talking about like, no, we're going to abstract this part of what you're doing and we're going to do it on another L2. I mean, it's wild.

Yeah it's totally wild. I mean, for me, tokens just like one very simple use case actually can be built around bridges, but of course bridges like most of the bridges implement something called Arbitrary Message Bridge. It's a protocol that allows to send arbitrary messages and this can be, can do anything basically.

Crazy. Do you think, okay, so as you were describing this scenario too, it starts to look a little oddly like Cosmos or something like these, like different things that are all connected. And currently there's like one central one, but do you envision, this is like, obviously no one can predict anything, but could you imagine that there's a future where there's these rollups kind of hanging off theorem and then they have their L3's hanging off them and then maybe the center of gravity shifts? Could that happen?

So it will never shift from a Layer 1 to Layer 2 or from a Layer 2 to a Layer 3 and that's simply not possible because the higher you go in the layer, the less secure it becomes. You always depend on the security of the underlying layer.

In your case, you're a full on blockchain with a token that

Exactly. So what we foresee is the future, which kind of combines the idea of Cosmos, of having multiple networks running parallel with the vision of having Layer 2's expanding the blocks, like the block space basically of the underlying Layer 1. So effectively we'll have multiple networks being securely connected via ZK bridges that can prove state correctness and that you're operating on the canonical chain. So this will happen and then those will have Layer 2's, Layer 3's whatever on top of it to scale like their specific applications.

So the other architecture that had this look a little bit would've been Polkadot with Kusama. Becuse I remember when they did those two.

Absolutely. I mean, ultimately the architecture has become extremely alike. So I think it's kind of a mix of horizontal scaling and vertical scaling that ultimately all of those systems derived towards to. I think it's great. It's like it's good to agree sometimes.

It's fascinating to see it all sort of converge in that way, but you can also see like there were different building approaches. Like how did these systems come to life? Yeah. Some were predefined and designed, some were sort of like specked and then picked up. I put Polkadot in the first one. I put Cosmos in the second and still figuring it out. And then Ethereum like, I guess Serenity had originally also predicted something like this.

Yeah. There were very vague ideas of like how the future should look like and has changed so many times. I think, yeah, Ethereum is kind of like, yeah, we just launched something and figure scalability out later on approach. Whereas Polkadot was more like top down, we solve scalability this way. And Cosmos was, I think, also more organic with the difference that it was always more app chain specific instead of having a generic environment for any kind of application operating on the same network.

Although a lot of the zones that are successful have had to sort of become full blockchains anyway because those connection points weren't really there for them to start using the other networks. Like now they're getting there. But yeah, it's sort of funny.

Like it's all converging to the same. So something that I also predict, another prediction. So kind of what we see is Ethereum is at capacity, right? For the last years actually. On the other hand we have like the concept of app chains. App chains basically have one network trying to have one application being run on this network. And that's it. And I expect actually instead of having this application sharding, we'd rather have user sharding. So user sharding for me is a bit similar to how you play computer games actually. You know, like we have different servers you connect to like the European server, US server. Even if you want to play something,

You're talking about like an MMO.

Yes.

Okay.

Yeah. So I'm talking about

not the CD that I put into

No, no. You play one of those MMO / RPG's right? And those can only scale because they have many different servers in different regions and makes sense because you always play with people. Well, many times you play with people in proximity. Well it's definitely kind of one way to solve this problem to have those different servers. And that's kind of how I predict also how blockchains ultimately will evolve. You have user sharding, you have like different Layer 2's. They will all have the same applications. But you will shard the users according to how likely they transact with each other.

And in a way it, so it's funny because it is sort of happening, but it's happening because of like the marketing efforts of teams. Like you'll have some L2's that just worse or like other blockchains that like went into a single market Japan or Turkey or something. And they happen to get a lot of users from that particular

It makes sense me how it grows. Like every user wants to have everything. Right. Like you're not moving to a city where you have only a bakery you need everything ultimately to feel like you can get around. And that's the most, the easiest to accomplish by having everything on the same network because you can have the same assumption of how costly transactions are. And I also like to compare to cities, you know, like depending where you live, you have different transaction costs and user sharding will also happen, not according only where people live, but also what kind of price they can pay for transactions. If you live in Manhattan, obviously you're willing to pay more rent compared to, I don't know, living in another city in US and that kind of, I like to compare blockchains to this as well. Like if you can be in the expensive shard where you can maybe do like high value transactions because you have a lot of money, you can pay a lot, but that's basically Ethereum. And then you have like other districts of the city that can be Optimism, Gnosis Chain, whatever, where people with less money want to transact or like other use cases are being developed, which just require different environment.

You had mentioned the Succinct bridge just before. Right. Let's dive into that in a little bit more, more detail. We did have Uma on the show as well. It's so funny this episode in particular is going to be like the list of other episodes that people should listen to. This is a very kind of like, you know, hub and spoke episode. But yeah. Tell us a little bit about the Succinct bridge. Where it's at. Is it already live and are you using it?

Yeah, so I will not dive into the details of ZK. Uma is 100 times better. This and you as well. But yeah, so effectively bridges connected different blockchains and they I think are considered the weak part right now of our ecosystem. Because they've been hacked a couple of times in the last years. And yeah, if you look at the bridge that's connecting Gnosis Chain and Ethereum we see it also as the most important part of the infrastructure because it allows these use cases that kind of allow to extend Ethereum with Gnosis Chain block space can be data availability, can be DAO voting, you name it. Right now today the bridge is still very simple. It's a multi signature wallet that is operated by many different independent entities that are validators on this bridge. They see a transaction coming in, they confirm there's actually mined on Ethereum, and then they trigger an event on Gnosis Chain. That's kind of how it works today.

How many multisig signers are there?

So we have 7 and a fraction of those has to confirm in order to actually make a transaction happening. And now recently just I think last week or the week before, we upgraded this with a new validator, which is Succinct.

Okay.

So what effectively is happening is Succinct can prove that a certain event happened on Ethereum by recreating the consensus of Ethereum on Gnosis Chain through a zkSNARK.

And when you say you've been added to, do you just mean like that is now a new part of multisig?

Yes. It's like one new entity in the Multisig.

Okay.

So we go step by step here. You don't want to go all in into the technology that is not in that sense proven. Of course everything has been audited. But it's a very new tech. So you have to be conservative and very conservative approach is just making them part of the infrastructure by being one of these validators on the bridge and being able to recreate if a transaction actually happened on Ethereum by proving the consensus of Ethereum on Gnosis Chain through Succinct and then using this as a confirmation on the bridge itself.

What are the other multisig agents doing then? Like are they also confirming something? Are they kind of doing the same action?

Yes, all of them. They're basically, but they are of course much less sophisticated. They're just looking. They just have an Ethereum node running and they see transaction happen. Someone deposited something, let's wait for this to be actually confirmed. I think 20 or 30 blocks or whatever. I think it had to increases for Succinct because like Succinct can only recreate the consensus. I think every 50 blocks or something. Don't quote me on the numbers. But that's effectively what the other validators are doing. And then if they see something was actually confirmed, they signal this and they sign off on this transaction and then you can trigger this transaction on Gnosis Chain.

Interesting. You're all of a sudden making me want to understand what the multisig, the regular multisig person or age like is it just a computer program running? Like it sounds like you'd have to do a lot of these. I always think of it as like a multisig versus like a human on the ledger.

No, no, no. That would be terrible. I think then Gnosis Chain would have no adoption at all. No, unfortunately we have just like computer services running. And that can watch the nodes incoming transactions wait and everything is automated and the reason why it's secure is because we have many parties running infrastructure independently and they're very close to us. So, and they are also very firm in security. So we know their systems cannot easily be compromised. That's why we consider the very secure system probably much more than many of the other bridges simply because the smart contract implementation of its very straightforward. Very secure. And yeah. That's why we think actually it's one of the most secure bridges today.

Now you're adding the, the Succinct and the ZK component

Exactly, but that's only one first step. Ultimately our vision is a different vision. So ultimately we want to move away from this old multisignature based bridge to a new bridge which we call Hashi

Hashi.

Hashi. Hashi.

Hashes?

So no, it's like the Japanese word for bridge actually.

Oh, okay. Cool.

So like one of our engineers, he's quite fond of Japan. So he coined this project Hashi and yeah, the idea of hashis, actually it's kind of also multisig if you will. What it does, it aggregates information from different bridges and allows applications. And actually a token bridge is an application of an arbitrary message bridge, which can be just built around, which basically allows to aggregate the information which is required in order to have these other applications built on top. And very simple example, we have multiple bridges already connecting to Gnosis chain today. So we have Succinct, Omnibridge, Wormhole, and a few others are running in a testnet environment right now. And if you think about blockchain information, we mentioned we need to extract information in order to understand that something happened on this blockchain, but ultimately the information is consolidated in one piece, which is the block header.

And all of these bridges are doing the same thing?

They don't

But don't they all have the same block header?

No

They don't?

No. So they work very simple like, like our multi signature bridge. They just have services running out. There was a transaction. Great. I confirmed this on the other network. That's how they operate. So they don't even look into the block header at all.

I see. Okay.

But the block header is, well ultimately everyone can still agree on. And to give you an example how we can do this, we can use contract on Ethereum that automatically when you trigger it sends the Ethereum block header of specific block, via Wormhole to Gnosis Chain.

Okay. Right.

So we basically, when we have this information in Gnosis Chain, we know it has, based on the security assumptions of Wormhole, we have this information relayed from Ethereum to Gnosis Chain. Now we can do the same using another bridge. Like Nomad whatever OmniBridge, Succinct. Right. They all can do the same thing and they will all relay in theory the same information, which the block header for a specific block and now on Gnosis Chain we can use Hashi to test if everyone reports the same. And if they all report the same, then we basically have additive security of all the different bridges.

That's cool.

And based on this block header, we can extract the information like, a token was transferred and allowed to build all the other applications on top.

Hashi though sounds like something you could use. Could you do something between like, I don't know, Osmosis and Ethereum or something? If they have like multiple bridges, they could create something like this?

Absolutely. So anywhere where multiple bridges are kind of connecting different ecosystems, you can use us to relay the same information, the block header to make sure we actually have trustworthy information. It's very unlikely that you can compromise all of those bridges at the same time.

Although wait, do they all have to be Ethereum based?

Well, they have to be able to connect to if you want to read

It's Consensus, it's not about the

It doesn't exactly, it doesn't matter that they connected to Ethereum. But they have to be connected to the blockchain that you want to actually transfer the block header from. If you want to build a bridge between Ethereum, Gnosis Chain has to be between those two networks. If it's, I don't know, Cosmos and I don't know whatever Osmosis or Ethereum then has to be on Cosmos and Ethereum and so on.

That's interesting. Yeah. Do you need it on both sides?

Well, it depends on how you want to do things. So for example, if you don't really want to go back to Ethereum, it is just about transferring information over from Ethereum to Gnosis Chain. Then the one way bridge is fine, that's fine. But in most cases, I guess you would like to have like if full duplex bridge, which allows to go in both directions,

Is it just a smart contract? What is Hashi?

Yes. It's just a smart contract. It's just a standard of how you can aggregate this information and decide on if you actually trust this or not. It's a very simple concept, but important because ultimately I think right now no one would trust any single bridge. But if you can add security of the different bridge implementations, then it becomes interesting. And on the one hand I think we kind of commoditize bridges here, which is also a bit tricky because most bridges care mostly about applications like token transfers, token bridges, but we just now care only about the block header, which is the most fundamental piece of information, but there's not necessarily like a business associated to it. And the idea is to build on top of this information a token bridge. In that sense, we commoditize it. On the other hand, I think for all of those bridges no one really trusts their security assumptions. So in some way it's also a way to gain trust again. In any of the underlying systems. By adding them altogether.

What happens if you spot one of them bringing the wrong header?

Very simple. Like in our case actually, bridge was hold and you would have a governance decision of what to do. Probably would probably disable this one temporarily at least.

Yeah. Yeah. But you'd have to like, like you'd have to tell, I mean,

Of course I mean the assumptions should never happen, right? Like if this happens and there's something security was compromised,

This is happening

Yeah. Yeah. Either technical flaw or just a security compromised whatever. And then it's better to just hold and see what's going on. Mm. Because otherwise, yeah, we've, I think people have lost enough money.

Yeah. Let's talk about maybe a new project or the new things that are happening. I don't know. Are they deeply related to the bridge? Not really. Right?

In some sense everything's related to the bridge because we have a lot of assets which are mirrored from Ethereum. For example, stablecoins, like UCC, USDt is not natively issued right now on Gnosis Chain, but it is coming from Ethereum. So you rely on the bridge security in order to use those assets.

So let's talk about the new project though. Yes. Gnosis Pay.

Gnosis Pay

Tell me about it.

Gnosis Pay is a new project that is launching at the time this airs probably. And Gnosis Pay is obviously about payments, which we consider one of the most important, but kind of underappreciated use cases in blockchain. If you think about the original Bitcoin White paper, the Satoshi Nakamoto actually described Bitcoin as a peer-to-peer electronic cash system. That was kind of, that's the title of the white paper.

Something you would use in the real world

Used for payments. Payments was the number one use case he had in mind.

It's funny when you say payments now though, we also think of on-chain payments, but here you're kind of talking about like,

Well, yeah, generally I would say regular payments, just like payments, how you would do it on a daily basis. And I'm not talking about like big transfers of money, but rather an everyday application and I guess today even the biggest Bitcoin maximalist would not argue for Bitcoin being used for regular day-to-day payments and of course.

There was a time in Berlin, what was it called? There's like this burger joint that would like, well

There was this Bar 77

Yeah

Yeah. Something like that.

Yeah, yeah, yeah. And he accepted Bitcoin

Exactly.

Eventually he probably didn't need to work anymore then.

He somehow run out of business. I'm not sure why. But anyways, so that was kind of the original idea of Bitcoin. Well, today we have maybe ideas like Lightning Network, but all of them would never ever scale to something that would allow actual like day-to-day payments to work. What are the reasons why this never succeeded? I think it kind of relates to one blog post that Vitalik also recently posted about the 3 transitions that need to happen in order to gain larger adoption, which is related to scalability through Layer 2's, which is related to user experience through account abstraction and which is related to privacy. Now every transaction is publicly visible and that's just not what you would like to have for your everyday payments. So I think we made a lot of progress on all of those topics actually in the last couple of years. And now we are, in my view, at an inflection point where technology is ready to actually enable those use cases finally. So we are able to develop Gnosis Pay as a Layer 2 with account abstraction enabled through Safe contracts on Gnosis Chain.

And what is it though? What is Gnosis Pay?

What is Gnosis Pay? So Gnosis Pay is the first decentralized payment network. So we are building a Layer 2 that is optimized for the use case of payments.

And by you mean like buying stuff in stores?

I mean, really everyday payments, like what you are using Visa, MasterCard, and whatever

So how would you do it though? Do you have your phone or do you have a card?

So as a kind of first product that is being launched on Gnosis Pay? We are launching a card. So it's like a debit card that you can also use with Apple Pay and, and whatsoever and it allows you to use funds that you are storing on Gnosis Pay. So you have your funds in a non-custodial environment but whenever you pay, you grant the permission to the payment processor to withdraw from your non-custodial account. Having your funds always in your own account instead of having Binance or your bank or whatever in between having custody of your funds

Because like there have been credit cards out there.

Yes.

Right. Crypto credit cards exist. Yeah. But usually behind the scenes, what's actually happening on those cards.

Yeah. It's very simple. I mean, Binance card, Crypto.com, all of those cards, they're just another interface to binance.com or crypto.com. So a user has to transfer the funds to Binance, a centralized exchange, and then you can directly pay from this balance and Binance makes 5% of the spread. The fee doesn't, like they say that it doesn't have any fee, but effectively when you start spending, they convert your crypto on the fly at the 5% spread. So you're actually paying a lot of money. Okay. So it's not even great from that perspective. At the same time, it's still very successful. There's definitely high demand for users to not offboard to their banks in order to start spending on a daily basis. And now what we do is move this from the centralized exchange to decentralized network where the user can have the same convenience, because I think that's very important. Like users don't want to change their behavior. You still want to use your phone or your card to pay everywhere, and you want to be able to pay everywhere. Like Visa, MasterCard, accept 80 million merchants. So it's the most distributed payment network. And I think in order for us to actually build a decentralized payment network, we have to build a bridge to the old payment network. Visa, MasterCard they're here for a reason and we have to connect to them in order to gain adoption.

And so that's the question. When you say you release a card, like what kind of credit card is it, is it just one of the more classic cards? Like it is, is it Visa?

It has to be Visa card because otherwise you could not use a Visa network and then you could not be accepted at 80 million merchants.

They've done that legwork on the ground to get

Right. So they, we still have to, at this point, we have to work with Visa in order to gain adoption and make sure our users have a great experience. At the same time we settled everything also on a decentralized network. So you have your funds on decentralized network and at the same time you can use all those funds anywhere in Visa. And the perspective obviously is to grow this network to eventually allow also peer-to-peer transfers between customers and merchants. Because that's the reason why it's so interesting. If you look at, we're always talking about removing the middleman, right. And we are always thinking only about Visa being the middleman. But if you look under the hood, then between the merchant and the customer, there's not only now the acquirer, the bank that acquired the merchant and the issuer, the bank that issuing the card for the customer. But now there's also a processor, aggregators, now there's Apple Pay in between. So there might be

Even a POS like the point of sale, the device

Point of sale, like it's insane. Like the amount of middle man in a regular payment transaction is completely insane. It's like, and every transfer that you do, it's probably likely that about 10 middle men are making a share of the profit. It's insane. And that's why it's also so insanely expensive to do those transactions. And we see the potential that over time we can start removing some of those and replacing them with extra decentralized technology. Because ultimately there should not be anything needed between the merchant customer being in the same shop doing a simple peer-to-peer transfer. What ultimately is obviously enabled to blockchain already

But in the past, what you've needed is each one of like the merchant and the buyer to have on their phones or in their space some sort of like equal wallet with the right tokens that they could transfer. And then they have to wait. And like make sure that the, you know, hopefully they're online and da da da. But here you're talking about just using a card.

Yeah.

Could you imagine like, what's the next step then? Like, you've sort of mentioned that you would even get rid of more of the middleman. Would you create then your own credit card and try to sell, like get that into the hands of people?

Well, I cannot talk about this

Just giving me a funny face there maybe.

Just kidding. So it's there's of course, let's say the short term plans, midterm plans, long term plans, short term and midterm is driving adoption. And yeah, we will have to rely on the existing infrastructure in order to gain a lot of adoption. And I like to bring the example of Skype. So I think I was also one of the first Skype users and I was really enjoying to do IP, voiceover IP calls with everyone else. While everyone else was still using landline, eventually Skype built Skype in and Skype out. Right. Allowed me for my voiceover IP account to call a regular landline and landlines to call into my voiceover IP account. And today, every call on this planet goes over voiceover IP. And I think we have to get to a similar level with financial transactions.

Right now, everything happening in the old world, a lot of gatekeepers and a lot of middle man making money. And we have been building in crypto over the last 10 years to technology that would enable something else, but somehow we forgot to build this Skype in Skype out experience, which allows to actually seamlessly connect between those two. And that actually allowed transition. And that's exactly what we are building with Gnosis Pay. So building this bridge that connects the old world and the new world and allows people to seamlessly interact. So eventually we can bring more people over and of course the more people on our side, the more likely everyone else going to join. I want in the end, Revolut to operate on Gnosis Pay. I want these neobanks to operate in Gnosis Pay, take advantage of the opportunities that we have been building in blockchain over the last decade but we need to make the first step, and that's exactly what this card is about.

I want to hear a little bit under the hood what's happening though. So like what I picture is that you have your account on. So when you, is Gnosis Pay a separate L2 or is it on Gnosis Chain?

So we going to launch on Gnosis Chain, but this will be a separate Layer 2 because Gnosis Chain is a Layer 1, it has the same constraints as Ethereum. So we have to operate in a different environment to really allow large scale adoption.

So what if you have something on Ethereum, you'd move to Gnosis and then move to the L2 on Gnosis Chain.

Right. But of course this can be made really seamless. I don't expect any user to do all these manually. And of course it would be also liquidity directly injected into the Layer 2. So it would be very, very straightforward to use.

Okay. But, so, but on that, you'd have an account.

Correct.

And then you also create an account for the card, or like, because how do you go into the real world? How do you connect those things?

Right. So this is where our bridge comes into place. So we build a bridge that connects to Visa at the same time it connects to your on-chain account.

Okay. What kind of bridge?

It's basically, it's like we implement the RP that is offered by different payment processors that allows us to operate in different parts of this planet. To set any kind of transaction. The crypto part is implemented through Safe accounts. Every user can give these different processes an allowance to withdraw from their account a certain amount per day. So it can be cryptographically insured easily through a smart contract.

Okay.

And now if you're going to pay, what effectively happens if user starts paying at the POS, we get the information via the processor that the Visa transaction has happened, the user has authorized this, and we can deduct this amount from the user.

So do you have like a Gnosis Pay on the L2 another account that's sort of like

Yeah, yeah, of course. You have like, you have basically a Safe,

So there's like 2 Gnosis Pay accounts on this L2, one is yours, one is the Gnosis.

It can be one actually it's just one account and we can just withdraw from this account because it's a safe contract.

No, but I, what I mean is one is the user, right? And one is the the processor

Yeah, that's right

On the actual network. And then that processor account is the one that's sending the signals directly to this.

Correct

Okay. And it, and I guess it's connected to a specific card.

Of course. Yeah. Yeah, yeah. So this mapping is done via our bridge. And I think what is, just to give a good example of how we can innovate and remove the man in this transaction, you probably have never heard of 3DS. So basically when you do large online transfers, sometimes you're prompted to do like a second factor authentication. And now instead of you logging to your bank account and eventually approving this, we can actually go straight and ask you to confirm with a private key.

Okay. Do you have to KYC? Actually that's the next question.

So if you want to use Visa, you always have to KYC but we don't see KYC as an evil process actually. We also try to innovate on this part. So there are two companies I want to mention. One is Fractal, we are using for KYC and they are building a new system called IDOS which allows user to own their data and easily share with other entities if they want to. So it's kind of like a distributed KYC platform where I can one time like get KYC information for Gnosis Pay, but then I can also use this and passport it to other services I want to use on this network.

They don't use ZK though, do they?

Not yet but it's compatible with it. And there's another interesting project I want to mention called Out DID sorry. You probably know those guys.

I think I've heard of them. Yeah.

So they're basically, they use the information if on the passport to create a ZK proof that Republic of Germany has confirmed this kind of information and it fulfills certain criteria like certain age or whatever without revealing this in particular to the entity requesting this information

But does Visa accept any of this yet?

Well the good thing is like we don't talk directly to Visa, actually. We are talking to the issuer and processor and we negotiate with them. Visa's actually insanely smart, like everything that Visa does outsourcing to other parties.

Oh wow.

So Visa is more like the brand. Saying like, this is how like you can operate on this network.

It's a standard body or something.

Like a standard basically. But many of the other things are kind of outsourced to other parties. And so it's us negotiating with them and educating them, but of course it's much easier than trying to change the overall like standard and I think they're very receptive to innovate on this as well and of course for us it's very important to have a great reputation, especially on this field because Crypto didn't make the best headlines.

True

In the recent past. Compliance is really important to us and that's why we make sure that every fund or all funds that are on the Layer 2 have to go through a KYT process meaning it stands for Know Your Transaction, by the way.

Yeah. I wondered if you just said it. Okay.

It is. It is. Yeah. KYC is know your customer. KYT is know-your-transaction. So we know the origin of the funds and we know they can be used in Fiat as well. And everyone operating on the Layer 2 can accept those funds and doesn't have to worry about that they could not off board

By the time this airs. Can anyone sign up for this? How do you do it?

Yes. I hope so. So you will be able, or you, you are able to buy those cards now. So you can go to gnosispay.com and you can go through a simple process to purchase one of those cards and then hopefully, yeah, the goal is in about 1-2 months, actually the cards are shipped and made available to everyone and then you can start using them everywhere.

Will this also spin out, do you think? Or is it spun out?

Yes, it'll definitely be also a spinoff. Simply because it's potentially at least a very big project that I think it's better to be executed in a separate environment as well.

What was the hardest part of making this happen?

It's not technology.

That's what I was going to say.

No. Like, it's even the short history, I mean, at some point they might be another session just on the origins of this because they're very interesting. And they're telling you a lot about the payment industry, actually. But no, the hard part is making people work together and making sure that an industry like payments where we talked about middlemen, where we make everyone happy, that is not easy because ultimately we obviously depend on this old infrastructure and we have to make sure everyone's aligned for us to scale and actually make it a widely accepted and successful product. And that involves a lot of politics behind the scenes. So the technical part is we have solved this technical part already for a couple of years now. But now I also realize like, why did payments not happen before because of this? Why is real world integration so hard is because you work with humans?

And I do wonder like the middlemen that you're cutting out probably don't like that you're doing that. And like, what do they do in today's world?

Right. So we're not cutting them out. That's a simple answer. Right now we are generating more revenue for them, so they appreciate it. And of course in the long term future, it's clear that there will be alternative payment schemes. So Visa is the number 1 payment scheme globally. MasterCard, the second biggest, and there are many others, actually, you would be surprised like how many alternative payment schemes they exist. Ultimately, we also would like to establish ways for people to actually do those payments. And I think there's a lot of potential and blockchains already by definition kind of the ledger that is needed in order to do those payments. Just have to make it more accessible and make sure yeah, it's actually distributed everywhere.

Sounds good. Stefan, thanks for coming on the show and sharing with us sort of a catch up on Gnosis, all of the different pieces, some of the spinouts and all of this new work you're doing on Gnosis Pay.

Thanks for having me, Anna. It was a pleasure.

Cool. I want to say thank you to the podcast team, Rachel, Henrik, and Tanya. And to our listeners, thanks for listening.