Can Kisagun, Co-Founder of Enigma, a cutting edge protocol that allows for ‘secret contracts’, smart contracts that allow for the sharing of secrets that can be used to power collaborative machine learning, talks about his journey as a spin out MIT with his professor, and the challenges of bringing a token backed protocol to life from the US.
We go deep into the possibilities of allowing a new more collaborative data economy, to move away from data silos, on big Impact areas like health, in particular situations like Corona and discuss new possibilities in the governance and coordination of distributed systems.
- His journey to working on Enigma
- Translating general technical problems to real world problems
- Tradeoff of privacy vs usability
- Multiparty computation – optimising for scaling in a commercial environment
- Why Enigma took a hardware approach
- What on chain transactional privacy means for them
- Comparing the trust Enigma brings to Web 2 business models
Jamie: Welcome to the Funders of Web 3 series by Outlier Ventures and me your host Jamie Burke. Together we’re going to meet the entrepreneurs, that backers and the leading policymakers that are shaping Web 3. Together we’re going to try to define what is Web 3, explore its nuances and understand the mission and purpose the drive its founders. If you enjoy what you hear, please do subscribe, rate and share your feedback to help us reach as many people as possible with the important mission that is Web 3. Okay, so today I’m really happy to welcome a guest from Enigma one of the co founders john Kerrigan, he and Enigma are somebody that we outline I’ve been working with for several months now on some really interesting projects. So it really exciting to have you here today to talk through Enigma and your journey as a founder,
Can: very excited to be here, Jamie.
Jamie: So to do a quick summary of your background like your co founder, guys this weekend, you both spent some time at MIT albeit you were at Sloan. He was at MIT Media Lab. So your own background was at Sloan. And prior to that, Northwestern, we did a Bachelor of Science in industrial engineering. You also spent some time Kinsey working in Turkish capital markets and media. And you are kind of a second time founder, I believe, previous project was st dust and XM check.
Can: Yes. That is that covers, I know 10 year periods, the concisely and given the name and the connection with McKinsey to Turkish markets.
Jamie: I assume you are Turkish.
Can: Yes, yes, I am from Istanbul. I grew up there and spend all my time until high school, then went to went to school in the US came back briefly. Then now I’m living in, in the US again, on the west coast, right? Yeah, San Francisco right now. Okay.
Jamie: So as I understand it, the kind of origin story of Enigma was that you met guy at the Bitcoin Club at MIT, where he was teaching a class. And you began to discuss and explore possibilities ultimately leading into enigma.
Can: So actually, anytime I was guys research projects, when I met guy, he was already working on Enigma, and I met him through the MIT Bitcoin circles. He was, I think, presenting his work and in Islam class, and then this was spring of 2015. The Enigma White Paper came out, and I’m not. And also in that guy’s primarily the author of that in, I would say the summer of 15. And, and then guy was teaching this class and in in spring of our second year, and both me and my colleague Tor were in that class. And then after that we just like, you know, kept in touch, stayed in connection. We both graduated in 2016. And he went on to do Enigma, I had a six month period where I was working on this project called EXIM. Chain. And then eventually our, you know, roads crossed again, in 2017.
Jamie: So it’s really interesting. Enigma is this kind of spin out from from MIT. Obviously, that’s a particular type of startup. I know, in the crypto world, there’s certainly in the US and from an investor perspective, kind of almost preference for spin out spec technical team. spinning out of these kind of top tier, Ivy League campuses, as a spin out from academia, your What do you think is different about that journey? And how did you find you were perceived by investors and kind of customers when you were when you first came out to market?
Can: Yeah, so, um, we had two major rounds of fundraising. One was in the summer of 2016, before I was fully involved. And I think at that point, Enigma was more pursued as an academic project from, you know, more traditional venture capital firms. And then there’s a second part which I was more involved in, which is the, you know, the token sale parts at that point. I think it was it was beneficial for us because a we had been doing a lot of work in the blockchain space, as an academic, as a project with academic backgrounds. So that was helpful. And the first one, I’m not sure. Like, I wasn’t part of those conversations. But I do remember that when I first joined, and we were working on when I was working on business development part of things, that we had this cool technology that we would go and tell everyone about, but at the time, people don’t really didn’t really know what to do with it. So I think in our case, it’s fair to say that in our early days, especially when we’re dealing with enterprise, we were this you know, kind of hammer looking for a nail situation, as most people criticise like no blockchain. A very common theme.
Jamie: Yeah. And I mean, I guess so we talk a little bit about the innovation itself. And obviously, the way that the project described itself has evolved presumably along that journey and something that is, I guess, one of the challenges when you solve a fundamental technical problem and it is You know, generalised or generalizable? The problem is how do you then translate that to be relevant to an industry or a particular use case? But generally, is it fair to summarise the work you’re doing as a blockchain based protocol specialising in privacy technologies that allow this kind of end to end, scalable baps? And at the heart of that innovation is what your to his secret contract? So that is allowing people companies to perform privacy preserving computation and to enable this secure data sharing.
Can: Yeah, that’s an accurate description. And yeah, I think the like, tying back to your previous question, that’s exactly what we do. But when we first came up with this, like, for example, people would have so privacy and, you know, and performance or usability Usually, you know, combat some, some trade off. And one problem that we initially had was like people would expect us to do highly performant things like running like a Hadoop database with, like privacy preserving techniques like, like multi party computation, and know that that was like, not possible still today is not possible. So, I think when you talk about like, you know, performance and generalizability, early on, we had expectations from, you know, enterprise partners that we’re working with, to, to have, you know, extremely high performance requirements, but also have privacy, which is, in the software privacy world, is somewhat unheard of.
Jamie: So there’s this kind of trade off that the world faces at the moment in almost every domain really, which is it’s kind of either privacy or usability and non normal environment is that true then blockchain, it’d be maybe worth just explaining the multi party computation piece. So I know you said that initially, you kind of went down that path when you were looking for taking the product to market trying to translate it into, I guess a language and a use case that people understood the pain point initially when you were speaking to enterprise, it was around multi party computation. Could you could you maybe explain what multi party computation is? How and how your approach with secret contracts is different
Can: multi party computation is is a software based encryption methods are similar to homomorphic encryption, where technically what you do is say I have a piece of data, we divide this data to multiple pieces and send it across the network. And in order for that data to become relevance, by that I mean that you know using the computation or Or like, you know, be displayed, we would need say, like 50 off 200 pipe spins in the network, obviously numbers are illustrative to come together and, and and perform this computation. So that’s that. That’s like the basic effect. That’s what multiple computation means that it’s similar to like, you know, Shamir secret sharing that people in the crypto world sometime use around constantly applications. But the whole idea is one party alone cannot see or cannot leak the data. So it has to be a group efforts to make the encrypted data be usable.
Jamie: and what was broken about multi multi party computation that needed a different approach?
Can: The problem is, is scaling of such software cryptographic techniques. I think one way to look at it is, one can either choose a specific application And work really hard to optimise that. So if you look into what z cash is doing with zero knowledge circuits is, you know, they have a specific use case. And then you know, they’re improving on that. But it’s it’s just, you know, sending and receiving coins in a privacy preserving manner. So, in our case, since we’re looking at computations, which is arguably a larger domain than just sending and receiving funds, it’s really hard to identify one specific use case and in really, really good performance in that. We were looking a lot of like in fraud detection, kind of use cases where, you know, Company A has some data, if they merge it with Company B, then they can do more effective for detection, but they can’t because of the sensitive nature of the data right now. And so that was our use case when we’re like having our enterprise for was still, even with that, like the computations that were involved. I don’t want to get too technical but like, required saving some name in memory and propagating that across every comparison. So just to give you an idea, like we were, at the time when we’re doing MPC, or this is first half of 2017, we’re much more performant than all academic papers out there, maybe like an order of magnitude. But still, like, you know, something like doing a linear regression would take us maybe I don’t want to give a wrong number, but maybe like, 10,000 times slower than like doing this with plaintext. And when you have, you know, enterprise customers who need to move fast, etc, like this, the scale issue became a big limitation for us.
Jamie: And so that’s interesting, you know, in an academic realm or theoretical realm, you know, you solve the problem. And maybe we could just talk a little bit about what was different, you know, how secret contracts sold for There’s problems in approaches around multi party computation. But then you had to somehow scale that in in a commercial setting. So maybe we could talk about what’s different, what’s different about the approach, how that solves for it, and then how you’ve gone about scaling that in a in a commercial environment,
Can: we have changed. So our mission has always been, you know, this the goal to push privacy preserving computation for better data security and better privacy. And we’re still working on that mission, but what what has changes how we achieve that so instead of doing multi party computation, which is a software based encryption method, we have taken a hardware approach, which has its trade off this hardware approach is called a Trusted Execution environments. This is a you know, similar to the fingerprint readers in our phones. That computers or servers have a separate enclave separate processor that is only accessible by like, you know that enclave alone. So it’s almost like when you send your fingerprint to your iPhone, your fingerprints never goes to Apple, but your fingerprint is only processed within this, you know, within this enclave. This also, this technology also exists on Intel chips that have been released in the last four years. So what we have done as a change in direction is we’ve learned to use this chip and build an operating system on these chips such that, you know, a user can encrypt data locally on the device, send it to this enclave in encrypted form. It’s only decrypted inside the Enclave and the security guarantees of the Enclave. Ensure that you The private key that can decrypt this input is generated inside the Enclave and cannot be leaked. So so that was that was a change in approach, obviously, like when you do this kind of change in approach, the computations become much faster you are now closer to the performance you get when you run computation plaintext data,
Jamie: right And so this is this kind of off chain computation. And the secret smart contracts are the things that then link what happens off chain where you can get higher performance with what happens on chain and can you kind of talk through talk through that a little bit more?
Can: Yeah. So, like what I described as like, you know, the trusted, Trusted Execution environment is a way to do privacy preserving computations. What I mean this can you know, work as a single server, this can work as a you know, as a blockchain. So, what we’ve done with the Enigma blockchain, we have a network have nodes that run these Trusted Execution environments. And that’s how you become a node. So it is not like an off chain solution. This is part of our chain. It’s it’s just the the nodes meet the special hardware to participate in the network. Okay.
Jamie: Understood. And I know that you’ve been working, linked to this around governance and the idea that smart contracts and some of the work and problems you’ve been solving for can also be applied to governance issues, more generally in the decentralised space, in particular auctions and the idea that you could have decentralised exchanges for computation, data, computation, and potentially even even dark pools as well. Right.
Can: Yeah. So there are a couple of interesting applications. I think. One thing that we have figured out as we know Dive into building this blockchain was privacy is not only relevant for handling sensitive data and you know, opening or allowing new applications to be implemented into the blockchain ecosystem, but it also creates significant usability gains to existing applications that we use in a decentralised world today. Some of these you touched upon, for example, things like governance, it’s important to keep, you know, the anonymity of the votes. So what’s, or at least during the time of tallying, in order to make sure that you know, people are not discouraged from voting if they see you know, the tally going in one direction. So the solution space that we had seen revolved around using this concept of a comments reveal, which means as a user, you would first commits a your vote, you take the hash of your vote, and commit to it. So no one could understand like what your commitment is. And at a later time, you would reveal your commit saying, Hey, this is actually have voted yes. And here’s my proof to it. And that’s how, you know, voting was supposed to be done. And if you think about user experience, like you don’t really want to go, and, you know, do a two time interaction, actually, like we’ve heard from projects working on this, that’s, you know, that issues of people committing their votes, but forgetting to reveal them and such and such. So, one thing that’s, you know, we hope to bring to the governance part is you can have accusing these secret contracts. You can have a single interaction from from the user, you can cast your encrypted vote to the Enigma network. And you know, when the time comes, the network can automatically decrypt your vote. Have the tally and show the results, you know, six years 14 or whatever, to the entire network without revealing user users vote at any time. So what we thought was like, oh, let’s focus on privacy and anonymity on the votes, we found that actually the value we brought was more on the usability side because at this point, you know, decentralised governance don’t really involve a, you know, critical votes, like your political orientation or whatnot. So that was like one area where you know, we see that privacy and usability, especially in the word free space go hand in hand and there are many more applications like this. This applies to the gaming space a lot we were in, in Waterloo, I believe six months ago, and there’s a team who was doing contract work for cheese wizards. I’m not sure if you remember this game, but it was by that collapse the the folks who are behind crypto kitties, and they will choose with this, which is a, like a glorified Rock Paper Scissors game that, you know, you play with wizards and their implementation because when you’re playing rock paper scissors if you go first and I know what you’ve what your move is, and I can easily make the winning move. So they also refer to this commit and reveal technique where, you know, you would choose your moves, you would submit them, and the other parts would choose their moves, submit them both in like no commitment form. And then once commitments are submitted, each player will have to reveal and then you know, you would calculate the winner and like a single game of rock paper scissors could take up to four hours or so. So team there took the game logic goes into the Enigma that, you know, we removed the requirement to the disco reveal, and they were able to cut the game length by like, at least 50% which means more people get to play there’s more activity and it’s like, you know, it’s helpful for This game. So these things actually go hand in hand a lot.
Jamie: So you guys have also been working on transactional privacy on chain transactional privacy. Could you tell us more about that?
Can: Sure. We call our transactional privacy efforts salads. That’s the name of the of the product, we call it salad because privacy is healthy. And salad is also another great example of where, you know privacy can bring usability to blockchain applications. What we do is slightly different than the zero know which implementations which require a user to create a proof and then and then verify that they have that proof in order to mix their funds between address a an address v. So again, it’s like a two step process. Similar to common trivial We replace that with a single user interaction using our network. So what the user does is do user would send in the recipient address that they want to see the funds go to, into a secret contract. And the secret contract would take in addresses from say 20 uses randomise them, and then submits those, those 20 randomise addresses to the contract that’s holding the deposits. And what this effectively means for the user is you do one transaction, and your funds are mixed. And this has been extremely helpful for us in pushing the conversation with Web 3 wallets, like meta mask and and my ether wallets, because our way of creating transaction privacy matches the way the users are used to sending funds. Again, this is going to be a high focus area for us going forwards.
Jamie: So when we’re talking about a lot of this stuff, it might seem quite navel gazing in that these are problems we’re solving for problems that have a niche, very particular to Web 3 in the crypto space and the rest of the world doesn’t really care about. But clearly, these are foundational problems that are critical to solve in the Web 3 space and around decentralised governance. So I know you guys have been exploring in a number of different domains to apply these technologies in. On the one hand, you talk about big impact research areas, you know how you can allow industries to begin to share data to run computation and presumably to share in the benefits of that and an equally you’ve also been looking at how you could begin To apply these technologies to the world as it is today in a kind of web two contexts, so what what you might do with Facebook data? So maybe we first kind of spend a bit of time talking about these big impact research areas. I know that Corona is very topical at the moment, and that’s something you’ve been actively participating.
Can: Yeah, no, I think. So, before diving into the details, I think one thing that we can say is like, either Web 3 or you know, web two, there is, you know, there is a trust model and what changes in workplace trust. So, like a very high level and basic example is, we can think about what Bitcoin has brought is taking the role of a bank, which would, you know, process me sending your money and distribute that to, you know, to tonnes of people anyone who wants to run a Bitcoin node and what that means is The trust model changes in that I used to trust the bank to do this. And also with like, with my data, now I need to trust the nodes to do what they’re supposed to do. And there are, you know, economic incentives built into it for them to be good actors. And then you also have to trust like, you know, these these random people on the internet with your data, I guess, where we see our work can bring value is around this trust factor. Who do we trust with our beta and to what extent now we’ll get into like, you know, like some of the data and privacy issues that we’ve been, I guess, seeing again and again in the past couple of years in the in the work to space, like we know that our data has been used without our consent, in ways that you know, we we do not condone, I think Facebook is a great example of that, and I I don’t think that 99% of us are aware of how say our cell phone location data is being used. So there is a fundamental issue about, okay, like, how do we use the data who gets access to it? Is it being used in the right way. And this applies to, you know, either institutions that control our data from like a user’s data being controlled by other institutions. But it also is, you know, is interesting when you have two institutions who, you know, can benefit from sharing data. So I think this is an interesting point, like how do we think about security of fire data and the corona example you gave, kind of fits like, you know, this realm where there is a lot of contact tracing, or I would say digitally, digital contact tracing efforts that are being proposed, with different privacy features, and We have lately volunteered to take some of our expertise in privacy preserving computation to build a better, a more privacy preserving way of doing contact tracing. And no, we are trying to work with some applications that have location data to, to know to run these analyses. What we’re trying to do there is not blockchain base, but we have a server that users can share their location history, and like diagnosis status, in a practice in an encrypted way. And in a constant this API is like secure data storage and computation platform where your data is stored. And once you like, you know, other people, like yourself, have contributed data to this API, the API without our access or our controls. Can can decrypt that data, run computations on it and give you insights in the context of safe trace, which is the name of our efforts. You know, you get to learn as an individual, whether you’ve been in close proximity with someone, or you also get some, let’s say, more macro analysis on how people who have tested positive have been moving around the city say, you need to go to grocery shopping. You can see which grocery shopping ago Google’s grocery store has lower risk based on people’s movements. And like all this can be done without revealing any user privacy. We were We were encouraged to this because we saw their governments, you know, initiate efforts like saying Israel that does this and has zero user privacy. So okay, you know, we can actually believe This using what we have, and provide better practice guarantees.
Jamie: So this is the big thing at the moment. I mean, it’s very Zeitgeist in that everybody’s talking about this trade off with privacy if we want to be able to respond to contain Corona is, is this approach that you’re talking about something that would augment, say what Google and Facebook are working on? Or does it? does it provide a distinct alternative?
Can: I think it’s Google and Apple. They just released something this past Friday, no word. So it’s actually a difference. Our approach to it. I think it’s really good that you raise it. The Google and Apple approach has a different security model. I think it’s very clean in that you and I say we both have our Bluetooth devices. If we’re in some close proximity, your device sends your device it to my device, and my device sends my device if your device and say five days from today, I get to sit positive then I I publish my let’s say, user ID or device ID, which is randomised say every day to a server. And you can query that server to see whether the user IDs that you’ve been close to have been tested positive. So I think this is a completely separate, you know, architecture. I really like the model because it is it has really good privacy design. If you can randomise these user IDs well enough. No information really leaks. However, this is only useful for no individual analysis or like letting individuals know after the fact there’s no location information tied. That’s why the privacy is very strong. But also this is not really helpful for say healthcare officials who want to respond to a high risk area, or this is not useful for me as an individual to take precaution. measures on like, you know, designing my routes, avoiding people with symptoms, etc. So it’s a completely different approach. But I think what’s important is, it’s important that big players like Google and Apple are coming into this market, because our approach focuses on location data. And if you think about location, data, location data already exists within big companies, big mobile apps, all that. And what we see right now is new applications coming up. And they try and get adoption. They know they’re trying to get users to build data, etc. Whereas, you know, if we’re able to tap into these existing companies, like say, like Foursquare’s yelps of this world, etc, then we already have this data and then we can bootstrap it. So, in that, I think it’s a good call to action that like, you know, big companies can can think about, Hey, can I let my user download their location data or can I use Let my user use their location data in a way that they can opt in to participate into this, you know, quite a bit, I would say, tracing efforts.
Jamie: So if we kind of build on that we zoom up a bit. And, you know, Corona is just the live instance of this, where you have a web two approach and a Web 3 approach. Are they distinctly different? Do you think these web two data platforms data monopolies can transition into Web 3? Or is that impossible? Is it is it Web 3 fundamentally different to to the business model?
Can: I think? So from a business model perspective. It’s a it’s an interesting question. The way I see it is I think these two worlds, at least from data side will converge eventually, because if you look into Web 3, it’s all about you know, self service. venti giving, giving users more control over things they do on the internet’s. And if you look into just a data aspect of what to, oh, especially with GDPR, in Europe, like, you know, technically speaking, I, as a user of an application, or a product, I shouldn’t be able to download whatever data I have. I mean, we’re not there yet. But that’s, that’s the direction that we will not head towards. So at that moment, if I’m able to use say Foursquare to, you know, get my location data, such that I can, you know, use it for COVID or whatever purposes. As soon as I have access to that data, then I can use that data in in into web two worlds. And there’s like, no way the the Oracle come into play. I think the chain link is exploiting some of these, these areas, but I think what to do data will eventually make its way into a three it will be used for certain use cases. And it will provide value for example, we can use like identity in the Web 3 was a big issue at how do you do identity how to do how do you verify people that if you think of identity as like, you know, different, let’s say, a large spectrum where like one end is a Hey, I need my audience to vote. And then one end is like hey, I need to prove that I’m not a robot then I think like using what to data souls a big portion of that spectrum where like, you know, we can use your Facebook data, bring it into an algorithm that say runs in, in a privacy preserving network like like Enigma based on that we could determine you’re not an actual bot, but a human being and then like, you know, issue like a non fungible, then you can use that non fungible to interact in the in the work field. Obviously, this is Not as safe as Hey, on junk. So again, I’m gonna vote in the next election. But you know, there is there is a there’s a role for work to data.
Jamie: And and in addition to that Enigma has a token, you’ve gone to great pains to, to bring that to market obviously, being predominantly based in the US, you’ve had to deal with the SEC, and that kind of high level of regulatory oversight that is not necessarily present elsewhere. But you know, you’ve still been insistent on the token being a fundamental part to the solution. Why is that? Why why a token? Why is it it’s so critical that you would go through the pain of, of working and dealing with the SEC to try to bring to market
Can: Yeah, so that was definitely a very tough process for us, and we’re glad that it’s behind us and we can find Focus on actually innovating in our space. And I think talking like when you look into a blockchain network where, you know, there needs to be network participation, we’re not talking about just an application, you kind of need to have a token or a coin that ensures that they are the right incentives for the network participants to act, you know, how they should be acting, and that’s like the broader theme of crypto, you know, crypto economics, because when you have a, you know, I just gave you an example of, of us doing this work for COVID-19. In a non blockchain setting, setting up a server, say that’s an IBM cloud with, you know, the rights trust model is very easy. What becomes hard is when you set that up, and you know, anyone who wishes to Participate can participate in this network, then, you know, attacks become more of an issue, you need to think about, you know, this honest actors and how they can try to interfere with what you’re trying to do. So, in a world like that, you know, the the token creates an incentive such that okay, like, you know, I have a, I have a reason to act in an honest way because I will earn these tokens, and if I only saw things, you know, that is that covers my opportunity cost, hopefully off of pipe fittings network.
Jamie: So is this a disincentive as well, right? So cost to malicious behaviour?
Can: Exactly. If you’re holding that I mean, within the proof of stake network, right. If you are, if you already have to hold all of these things to attack the network, then technically speaking, you’re attacking the network would be detrimental to you and to your holdings as well. So a token needs to exist. It’s a network That is truly decentralised.
Jamie: I don’t know how much you can talk through this. But it would be good to understand how you manage that engagement with the SEC. I know you’ve now transition to to cosmos, and presumably the network, it very deliberately has certain characteristics in its in its design and execution to allow you to have a functioning token. Can you can you talk us through the thinking there?
Can: Sure. I think one thing that we learned after this, you know, after this process is we need to be, you know, very, very careful in the decisions that we make, and we need to make all decisions in a decentralised fashion. So, as you mentioned, we launched a Cosmos based book chain, approximately me on two months ago, the launch of the network happens With participation of over 20 different stakeholders, including outlier ventures, and the new network has a native coin that we call secrets. Secret is, you know, the governance and choose governance of the Enigma blockchain. So all the governance decisions are made with secrets in a decentralised way. We have changed our role compared to like, I would say maybe six months or a year ago, where, at this point, we are a development team that builds new features for the blockchain based on what we see fit. But you know, whether what we build gets implemented and will be determined by you know, the governance decisions that secret holders do so at this point, we’re just building and proposing and any kind of execution of what we propose and adding this in To the to the Enigma network is solely based on the discretion of secret holders. And like other folks can also propose network improvements, and that can be voted upon. So I think what I’m trying to say is we have a much more decentralised approach than what we had six months ago. And we’re just one of the many stakeholders in the ecosystem right now.
Jamie: And it sounds like a logical pathway anyway, right? I mean, I imagine this is something that you, you were aspiring to realise. Anyway. It just so happens that perhaps it’s it’s had to be accelerated?
Can: Yes. Yes. We always wanted to think about decentralised governance. And I think two things have played into it. Obviously us using Cosmos technology helped us implement this in a much faster way. And obviously like, you know, When you have major things going on like force measure, you have to be reassess certain, you know, certain options and make decisions that are not only not only reactive, but also are for the better for the better health of this project going forward.
Jamie: So obviously, in that use case, the karate use case, you’re talking about how individual citizens can, can share information. I know the outlier labs team have been working with you on how that could be extended to enterprise and corporations working on something called the cross enterprise collaborative analytics. Could you could you explain a little bit more about that?
Can: Indeed, as,Yeah, as we mentioned, the secure data sharing involves different parties sending data into the secure storage and computation platform in the case of safe trace. It’s, you know, patients But also in the in the realm of enterprises could be different companies. This could be different business units. The work we’ve done, actually was an inspiration for for safe trace, where, you know, we create this weekend environment for two telco players to share location data in a privacy preserving way, such that they could drive business insights as to where to, you know, build cell phone towers and for better coverage. And we also see this this use case being or this technology being very relevant to solving issues around frauds, whether it’s user fraud or or driver fraud in the share, right economy and delivery economy. So this is a very promising area.
Jamie: Yeah, well, look, thanks so much for your time. We’ll have to wrap up They’re I know, the outlier ventures labs team have been doing some work with you on something that’s called the cross enterprise collaborative analytics, which was again around data sharing. So people should check that out. But I think it’s going to be really interesting watching this pathway to decentralisation increasingly having a roadmap, both informed and potentially even executed on by the by the wider community. So thanks so much for sharing your your journey with us and good luck.
Can: Thanks, Jamie. It was a pleasure to be on your podcast.
Jamie: If you enjoyed today’s podcast, please make sure you subscribe, rate and share your feedback to help us reach as many people as possible with the important mission of Web 3.