Fireside | Building Configurable Privacy for Real-World Payments

Thank you guys uh for attending here. uh we we we would love to expand with Dmer uh um on stellar how we can actually you know uh be configurable privacy for real world payments and I think stellar has been very like forefront you know uh in that kind of innovation and you know like early January uh you guys uh basically uh made a protocol upgrade to the to the network that enabled basically the primitives uh for uh those kind of application and capabilities actually. So um uh I would like to explore how like you you recommend people actually to move from okay we do have now the primitives to enable those kind of products to exist on stellar to actually have those kind of uh features in place. So how maybe a payment operator that want to deploy their you know product on stellar can actually start using those things.

Sure. So I think um the on the topic of privacy um our approach is very uh modular in the sense that it's really important for us to keep stellar the L1 as open neutral and transparent as possible. And so I think recently we've seen uh you know there's a lot of permissionless a lot of permissioned chains that keep coming up and this is a nuisance in the world of crypto that keeps coming up every couple of years and uh now it's kind of like in service of of privacy. But when we talk with institutions that are coming uh to build a blockchain they're really interested in the openness in the auditability and the the verifiability of the chain. And so it's very important for us to maintain these properties as we introduce uh privacy.

And so our approach at stellar is uh is uh on the L1 itself on the protocol. We want to make sure that it's completely open. Anyone can come and build. We recognize that there is um uh there's a spectrum of privacy and so you can have uh privacy that relates to different jurisdictions, different use cases, different tech. Obviously, we're talking a lot about ZK right now, but there's a lot of other uh uh potential tech both now and in the future.

And so, we want to make sure that Stellar is an open network and uh and stays that way. Um so, uh for that, we launched uh we upgraded the network to protocol X-ray a few months ago. X-ray uh delivers a set of cryptographic primitives that allows you to uh to build your own privacy protocols on top of them. So that is the BN254 curve beside and hash functions uh kind of like really enriching the set of primitives for ZK. Um so that's on the protocol itself.

Uh with that said, we're also doing a lot of work on the uh ecosystem development and partnerships front to make sure that you know we're just not we're not sitting around waiting for people to build privacy, but we're collaborating with industry leaders such as yourselves, such as Nethermine, such as Open Zeppelin on bringing in uh a variety of solutions that have a variety of uh you know privacy guarantees and a variety of administrative controls. That's that's makes uh totally sense. I think you you you guys launched also on your own like stellar private payments to show actually those capabilities and uh yeah I think on our end at boundless I think this is a uh an approach that we are following as well you know with our uh upcoming modular framework the boundless digital stack where like we do believe that basically institutions and we are like testing actually during these events that need a credible layer actually like stellar to deploy their capital and they need those capabilities is in a modular fashion. So they need to be able to integrate seamlessly on their existing infrastructure as well as maybe like uh also have some end to end vertical solution. And I think the modular strategy is actually the thing that you know like in general in defi in crypto actually uh worked out.

Uh cool. Um I think maybe uh the next thing that I have for you uh Tomer is um in practice configurable privacy what what what actually does it mean like what I should like you know shield what I should should actually keep transfer because of course institution doesn't want to like broadcast you know the values their strategies on chain or like who are the sender and the recipients so but how they do decide you know what they can disclose and actually how they stay also compliant you know on operating those kind of you know uh applications. Yeah. So I think when we're talking about configurable privacy there are kind of like two main dimensions uh that we think about in terms of of the configurability. One of them is what are the actual privacy guarantees uh that you get and the second one is what types of administrative controls we have in place in order to be uh compliant.

And so on the guarantees I think specifically with payments uh the main uh properties that you can think about are uh transaction details that is you know what are what's the actual amount uh that I'm sending and then there's the identity of the senders and receivers and so what we're seeing right now is that uh we're seeing a lot of work done on the confidential uh token space we're also collaborating with open zeppelin and the confidential token association on this Uh this is um uh basically an ability to hide transaction details while maintaining senders and receivable and receivers uh visible onchain. So this is really good for use cases like payroll for example where the identity of the players is well known. So everyone knows that I work for SDF. Uh what we're trying to hide in a use case like payroll is the actual numbers. Right?

I don't want uh a payment to expose uh both you know my my existing kind of like salary but also uh the complete history of um of uh of transactions that I've committed. And so uh that's something that we're doing on the confidential tokens front. We're also seeing a lot of work on the privacy pools front. privacy pools adds uh these are payment protocols that also add um kind of like a mixing um component to it so that you uh completely obiscate the senders and receivers. So this is kind of like uh better in like kind of like the general privacy sense and it it also adds anonymity but it's kind of like more complex and so um there's also uh more compliance variables there because you want to make sure that if I'm in a privacy pool I'm not in a privacy pool with a bad actor because if I am I am with a bad actor then it might uh taint the funds of others in the same pool.

So this is on the side of the privacy guarantees on the side of on the side of administrative controls we're seeing a variety of administrative controls and because the world of compliance is not really caught up to the world of private blockchains or privacy on blockchains we're we're not really sure what compliance means yet. And so what we're trying to do is to build a menu of administrative options. And so um some of the things that we're looking at are selective disclosure. This is something that we've seen a lot with view keys and so the ability of a user to be able to uh to prove their source of funds without revealing it to the entire world. So kind of like ad hoc uh proving source of funds.

We're looking at things like non- selective disclosure. So say for example that uh that a a law enforcement agency wants to get access to a specific record. The question is how do they get access to that record with uh minimal uh access to the rest of the system. So um you know for example through global audit keys and things like that. Um and then we have uh things like association sets nonassociation sets with our which are I was talking about the issues with privacy pools and mixing with elicit funds.

So association sets are kind of like a a way to gate privacy pools. Um, so these are some of the things we're seeing kind of like some more options arise around like clawback functionality and we just want to make sure that we're not putting all of our eggs in one basket and so we're allowing for different types of solutions with different privacy guarantees and different administrative controls to coexist. Makes sense. So what I what I hear is basically that uh depending on the actual use case you want to have let me say a diverse spectrum of like privacy options and uh and that's actually the value like you know of the primitives that you have been enabling and like products as you mentioned like that we are building as well other players association actually that try to standardize you know uh that thing okay that's that's uh make totally sense and if If if you want to uh focus a little bit on like long-standing institution especially banks and you know traditional finance actors uh what they're like uh what are the tools that they can actually start using uh today with so what are like the the approaches that they can use to uh stay compliant but start actually to deploy capital on ster or in general on chain you mentioned like uh ASPs is is like a similar concept that they already have uh in their like compliance tools but yeah what are maybe others approaches that institutions are exploring?

Sure. So um you know we just put out a project called Stellar Private Payments which is a collaboration with Nethermind um and that's basically uh the idea there is having like an open- source package that financial institutions can get acquainted with. They can deploy their own version of it. they can modify that and we're building uh this menu of configurable privacy options and configurable administrative options right into this and so that's something that people um can use uh right now we have a long road map uh ahead of us and so uh we're really looking forward for uh feedback and so uh and I'm already hearing from other projects in the space that are starting to use that as a baseline for deploying uh their own privacy solutions. Um, I am curious, you know, uh, you know, you guys at Boundless are also in the process of of building your own shielded vault.

Um, curious to to hear how you're thinking about, um, you know, administrative controls, especially around the idea of building something that is, uh, flexible enough for different institutions working with different regulatory requirements.

Thanks for the question. I think uh uh our approach as we mentioned earlier is is very modern in this sense. So uh we basically have onchain capabilities to uh define inclusion set or non-inclusion set. So basically that you're a good guy or at least like you're not using uh bad guy uh monitor. But also an interesting approach that we've been following and I think that ZK is uniquely enabling that is that whatever compliance logic you do have offchain you can bring it actually onchain and verify that payment is compliant and operation is actually compliant uh through ZK proofs.

So uh we've been integrating with like uh tools like theorem labs to comply advantage to the traditional you know tool belt that institution actually do use and be able to basically uh when you do a a movement on chain you do a payment you do like an investment you do like some management that basically those checks are actually valid. So you don't have to as a institution you can actually use newer tools that maybe are more like blockchain native or uh add some more capabilities to the uh compliance that you need to do on the other end you can just bring it on chain whatever you use and I think that's that's very important for C institution because generally we see with other approaches more like hey you have to launch your own chain okay or you have to deploy on our own chain or you have to use a specific framework. That's a very important front cost for them because they need to learn something new or maybe they have to spend like six to seven figures to actually run the chain and also track capital there. Instant here the idea is that okay wherever you want to operate okay our case like stellar uh but also other ecosystems uh you can just use this framework this product deploy the things where where where you need and just start using without any sort of like big front investment. We we already know that especially institution have let me say longer cycles in general of product development and you know sales and stuff like that.

We want to basically reduce to zero the entry barrier there because it's really like hard you know uh blockchain in general those kind of things for people that don't use them uh day by day and so I think that UX and on boarding is definitely something that you need to keep in consideration when you like as a pillars to the product development.

Very cool. And you know you talked a lot about ZK and I think we had a little talk outside about this. Um it it does seem like zero knowledge cryptography is getting all the uh the spotlight right now in terms of the actual privacy tech and ZK is is uniquely suitable for for payments specifically because they can be easily expressed as as as a as a as a set of uh of of ZK proofs. Um the question is what is the is this going to stay with us? Uh how are you guys thinking about the future?

How are you thinking about non-payments use cases? you know, we're talking we're thinking a lot about uh private DeFi, about private trading. Uh I don't think it's going to come uh you know, this year or or in the next few months, but this is around the corner and ZK is not necessarily always the right solution for it or or is it?

Yeah, that's that's a that's a very good question. Thanks for for for asking. So uh timelines uh you know uh can change depends on a lot of things but uh so two things I want to say. So the first one is that definitely ZK is the first candidate as a technology uh to do these things but we cannot reduce to ZK to be the only thing for privacy preserving or let me say confidential uh uh finance or defy here. There might be uh a spectrum of other approaches.

Why do you separate confidentiality and privacy? What are the different terms mean? I

I I think that uh I tend to associate to privacy uh to like full privacy. So being able basically to stay fully anonymous in that sense whereas for confidentiality uh that's something that basically you decide what you actually disclose and uh uh that's for me more like uh configurable privacy in in in that sense. So like we we've chatting before like you know uh being able to selectively disclose to third parties about those operations when not maybe not always but when you need so be compliant with you know the legis legislation where you operate the regulators or maybe other third parties. So that's what I I mean but getting back to to the topic about confidential defy, I think that um we can definitely operate like on existing protocols about lending or like taxes um within using privacy pools and uh I think the uh technical challenge that we do have today sometimes is how we can enable composability across those use cases. So for the single use cases, so if you want to do private swaps, private you know uh uh private decks or even like private lending uh confidential lending you you can definitely do the challenge that we are uh exploring when we are basically we have a stream of research about composible proofs actually ZK proofs is that how you can you know like have composibility across because I think that composibility ultimately is something that DeFi innovated a lot.

So being able to compose those operation naturally atomically is a good value. So being able to recover even in a confidential environment is actually something that uh very important to to achieve. I'd say that probably like uh by the end of 2026 we we going to see something like that we we'll work hand to hand and we'll able to compose.

Very cool.

Yeah. Okay. Uh I think uh just uh uh be respectful of time I I have a uh a final question for you uh Tor which is uh traditional institution so far operating in a more like let me say uh opac fashion. So all the software that has been running has been always proprietary and like it's very hard to like understand what's going on and uh also from a security perspective like you you don't really know if you know there is a bug or there's what's the security risk there and that's actually a huge cost for you know institutions uh and uh I think that blockchain technologies that are especially invested a lot in in being sure that all that kind of technologies and the most important stuff are actually open. So I'm wondering what's your take about like having those set of techn compared also to other like players that are trying to solve this problem to be like an open framework and an open space where people can actually you know dive deep and basically verify on their own.

Yeah, I would say that you know uh blockchains in general and obviously privacy specifically these are uh systems for interoperability and part of being able to interoperate with other actors is that there is a very clear set of rules that everyone agrees on. And when we're talking about blockchains that are open participation, not like some of these other kind of like permission chain solutions, but real true um um open permiss open participation chains. The only way that we can achieve this complete alignment where everyone knows the rules is if they're completely open source. And specifically um specifically uh for privacy this is even more important because ZK can be uh so opaque sometimes that without having this clear visibility it will be very challenging to get confidence in this technology.

Fully agree to answer the same way I think we are actually uh on time. I I'm wondering if maybe someone from the audience have any any question for us. May we stay here and so I see a a question there. about institutional uh we're speaking a lot about institutional permission permissioned interactions specifically but I was wondering what's in it for permissionless individuals um you know uh not only the chatfi institutions is there anything for smaller uh scale users to take take into account here

yeah look you know for for 10 years in crypto we've been dealing with like the the ideological version of of privacy and and peer-to-peer transfers. There's great tech. It's available. It's hardly ever used because it's not very useful, right? And so the ideological value does not necessarily translate to to monetary value.

And if we want adoption uh for everyone, not just for the institutions, you know, at the end of the day, you know, if if an individual wants to be able to uh receive a payment from the company that they work with, if they want to be able to offramp that into cash in their hands so that they can actually use their crypto, um for these things, you actually need the institution the institutions in play. So, you know, if we have like this, you know, beautiful or serendipitous uh um um you know, closed uh circle of privacy, that doesn't help anyone.