07725 160903
A transcript of the Future of Finance interview with Kåre Kjelstrøm, chief technology officer at Concordium.
Dominic Hobson 0.14: Hello, I’m Dominic Hobson, Co-founder of Future of Finance. My guest today is Kåre Kjelstrøm, Chief Technology Officer (CTO) at Concordium, a Layer 1, proof-of-stake Blockchain with an integrated counterparty identification functionality, one of whose founders is Lars Seier Christensen, Co-founder and until 2016 Co-CEO of Saxo Bank. Kare, thanks very much for joining us.
Kåre Kjelstrøm 0.38:Thanks for having me.
0.42: What is the history of Concordium?
Dominic Hobson 0.42: Now, I’ve mentioned one of the founders, but what is the history of Concordium? The strategy? Are you Danish? Are you Swiss? Are you something in the middle?
Kåre Kjelstrøm 0.52: It’s a very good question. So back in 2016, when Lars had ended his Saxo Bank adventure he was you starting to look at what sort of the next thing [was] for him. And he liked blockchain but he also had this … When he looked at it, he was like, `This is fine, but this is also [a] completely unregulated market; this is a new thing that’s coming up, and it completely transcends everything that’s going on. And, obviously, this, this is not going to last, right? At some point in time regulation is going to take a look at blockchains. And something will have to give.’ So he felt that it was a super-interesting space to go into. And he kind of wanted to potentially even make his own blockchain. So how to go about that? How do you build a blockchain? Well, what you do is – and build a blockchain that’s good and safe, by the way, right? – what do you do is you call some people who know crypto. And it so happens that at the University of Aarhus, here in Denmark, where I am, there is a group of cryptologists at the university who know a lot about crypto and actually some of them came up with some of the foundations for cryptography as well hash functions and other things that are like key to blockchain that was actually invented here. So he called them up and he said, `I want to build a blockchain. Can you help me? And that was sort of the beginning of a very fruitful collaboration with that department over there. So Concordium co-founded a group called COBRA, [or] Concordium Blockchain Research [Centre] at Aarhus University with the crypto group over there. And over the years since 2017, and up until now, we’ve been collaborating with COBRA to design the science behind the blockchain, basically. And we’ve sort of published, I think, around 70-plus white papers at this point, that detail all kinds of aspects of building a blockchain, which has since been turned into the Concordium product. So that’s really the story of how we began. The company is a Swiss entity, but in so many ways it’s also a Danish entity. I’m in Denmark, many of the engineers are here. We also have engineers in Zurich. And so Lars is Danish, but he lives in Zurich. And so, you know, it’s kind of a Danish/Swiss company in that sense, but we are a Swiss entity.
Dominic Hobson 03:16: And your own background isn’t actually in blockchain. So what’s the … (a) what attracted you to this project? And what experience and insight do you actually bring to building this layer 1 blockchain?
Kåre Kjelstrøm 03.28: Yeah, I have a master’s degree in computer science from the same university actually, but it’s somewhat dated – 24 years ago is when I took it. And since then, I’ve just been, I’ve been working in Denmark, in the United States, and in various small start-ups. I had my own company, Eventually, back in 2014, I got a chance to work at Uber at a time when the company was only some 400 people. So I was on a growth journey with Uber from 2014 to 2020, where the company literally grew from those 400 people to 27,000. And the reach of the company grew from being just United States, and I think Paris and Stockholm, to pretty much the whole world, right? So it was a tremendous journey for me in terms of learning to work in a hyper-growth environment [and] how to build scalable software. Eventually, I was managing the storage team at Uber, and once I finished that, I moved on to a company called Cloud Kitchens, which is Travis Kalanick’s newest company. So I was managing storage computer network and site reliability engineering at that company for almost two years until Lars and Lone [Fønss Schrøder], my CEO, called me up and asked if I wanted to come on board as the CTO of Concordium. My learning from both Uber and Cloud Kitchens is really about how to how to build high performance teams, how to build scalable software, how to make sure that you bring something that runs 24/7/365. And how you productionize something. And I think that set of experiences is something that’s relevant for any kind of product that you build, and in particular also blockchain. This is an operation we will never take down. We don’t have service windows. We don’t basically shut the blockchain down to just, you know, service it, right? And it cannot … We cannot be in a situation where the blockchain crashes. It’s simply not an option. So bringing all of that, and my experience in scaling teams and building them out, I think is key to my job here.
5.28 Concordium works closely with universities. Why?
Dominic Hobson 5.28:So you bring the experience of riding a rocket ship and making sure it doesn’t crash. But as you stated at the outset, the organisation is still working very closely with people at the universities. What’s the advantage of working with university employees as opposed to people who’ve had experience akin to your own?
Kåre Kjelstrøm 5.54: I think you kind of need both. If you look at this – the blockchain space – a lot of the projects out there I actually would call tinkering projects, right? Someone decides that, `Hey, Blockchain is probably cool. Let me go down to my basement and tinker a little bit with it or let me clone Ethereum and set my own thing up.’ Best case, you get something that works. And you might even get something that gets traction, and you might even sell a lot of tokens. You might even get rich. And then the blockchain crashes, or your Stablecoin isn’t as stable as you thought it was because you’ve been doing something in background. There are many examples of this. And if you build on science, which is what we’ve done, you have a much, much better chance of building something that will actually be robust, secure, safe, and can actually fulfill the promise of a blockchain – but making it such that it’s scalable. It’s a whole different ballgame, right? Making it such that you can just, basically, handle potentially any number of accounts, any number of requests – that you can scale beyond the sort of the basic foundations and get to, like, high transactions. It’s a whole different thing. So for that you need different kinds of experiences. But it’s also a blockchain world. And it’s also crypto. So you need, you know, basically both worlds in order to make this happen. I think that’s what I can add to what we already have. We have a ton of smart people involved that know everything about crypto. And I come from a world where it’s about really leading the market, scaling fast, building up, never being down. And I think that’s a great combination.
Dominic Hobson 7:28: Now, one of the things that really caught my eye about Concordium is this integrated digital identity functionality. And I think I know why – I think I know why it’s important and have my view on that. But why do you think that is important to the success of a layer 1 blockchain?
Kåre Kjelstrøm 07:43: So that’s actually the unique property that we have in Concordium. In order to even open an account in Concordium, you have to go through this identity verification process. And it’s, you know, you take a picture of your passport, you do a selfie as part of the process. And then there is a system in the background that does a check, that checks the passport looks right and that your picture looks like the one on the passport, and then we can issue a digital ID. This is done by external identity providers. So it’s not like Concordium does it but these are partners that we integrate with. Once the ID is there, it’s in your wallet. And every time you do a transaction, you move our native coin (the CCD) or you buy or sell something, that transaction will be recorded on the blockchain, and there’ll be a reference – an encrypted reference to a pseudonym that basically identifies you, right? That pseudonym is then used as a key over at the identity issuer. And if you have it, you can unlock it. But in order to do that, you would need some keys and those keys we have put with some law firms that are that we call `anonymity revokers.’ So not a single person can actually do this. This is something that can only be done by everyone involved. And so if law enforcement comes knocking, we can point them to the anonymity revokers and they solve it. This is actually what we believe is key to creating a platform upon which you can build applications that comply with regulation. There are so many examples of things going on in Europe and the United States right now, in terms of trying to figure out how to deal with Anti Money Laundering (AML) problems, counter[ing] terrorist financing (CTF). There is the [Markets in Crypto Assets] MiCA Act [in the European Union]. The [Securities and Exchange Commission] (SEC) in the USA has even created a task force to deal with blockchain and figure out how to handle this. There’s the EU sixth directive on AML [the Sixth Anti Money Laundering Directive]. and CTF. And so we believe that in the coming years we will see the established world of finance and the established world of government come to the world of blockchain and say, `You know, it’s great what you’re doing, but you need to start playing ball, right? You need to start allowing for regulation to come in and deal with what’s going on here.’ And for that an ID is key, right? So if you can resolve who actually did something, then, you know, you can find out who did it and potentially wouldn’t need other companies to go and analyse the blockchain and find out maybe it’s some person or someone else.
Dominic Hobson 10:20: So it was really the imposition of the [Financial Action Task Force] (FATF) recommendations on customer due diligence, if you like, back in 2018, which drove you to incorporate this feature. It’s interesting. I was looking at a FATF survey of progress on the implementation of AML/CTF in dozens of jurisdictions around the world, and hardly anybody, hardly any country, had actually made any serious progress on it. And there we are, you know, almost exactly four years after that [obligation] was laid on the industry. And I sense as, you’ve indicated, the regulator’s getting a bit antsy about that. And so it probably will be a tremendous sales feature for you as you go forward. But can I tease out a little bit what those partners who work with you [are doing]? You’re working with a bunch of digital ID vendors. You’re also working with these, these law firms, which I think you call an anonymity revokers. Can you explain in a little bit more detail how [and] what each of those partners actually bring to the party? Presumably, the vendors bring that photograph of a passport component. But the lawyers are fulfilling some slightly more important role. They’re revoking people’s anonymity, presumably.
Kåre Kjelstrøm 11.36: Yeah, to some extent, right. So the important thing to understand is that we sort of split the knowledge of figuring out who was doing the transaction between multiple parties. So not a single party can go and figure it out. The anonymity revokers are really law firms. And we’ve given the law firms a ledger with a key on it, and that they keep in custody. They cannot use that key individually. If the other keys are brought together, then the combined set of keys can unlock any transaction on Concordium. And what that would reveal is a pseudonym, which doesn’t say anything about the person behind it. It’s just basically a random string, plus the identity issuer that issued the ID that corresponds to that particular pseudonym. Now, with that in hand, you can now go to that identity issuer and say, `Tell me, this particular key here, the string, who was behind that?‘ And then the identity issuer can go ahead and look up the string and then find the ID information because they’re storing it. The identity issuer cannot use the pseudonym for anything. They have the database of pseudonyms to ID but they can’t really use that because they cannot decrypt the pseudonym on the blockchain, right? So that’s really the way we sort of divide this. Is it perfect? Not really. We want to make this much more decentralised. We are looking to do multi-party computation (MPC) on the keys and basically abstracting it away from single entities. And the same thing can be done for IDs and other things. So we are only at the very beginning of how this should work. And this is a very centralised way of doing it, to some extent. It’s still, you know, [done] in a way where not a single party can resolve it. But we’re looking to improve on this particular aspect of the chain.
Dominic Hobson 13:26: I’ll ask you about your views on digital identity in general in just a second. But first a couple of points of detail. Can people choose their own law firm? Can they choose their own digital ID vendor? Or do they have to use yours?
Kåre Kjelstrøm 13.42: Right now, in terms of the ID vendors, these are external companies, obviously, that do KYC [Know Your Client] for other customers than us. We have just signed up with two and we are looking to make an infrastructure where it is easy to onboard any number of identity issuers that live up to certain standards. So, at this point, you have to choose between the two we have signed up with. Like, again, I said it’s not perfect, and it’s a thing that will definitely be evolved as we go forward. In 2023, on our roadmap is the ability to open up for any number of identity issuers, and even not just those that look at the legal identify, but also potentially companies issuing, you know, employment IDs, or your local chess club issuing your card or loyalty cards or other things which you will then be able to keep in your [digital] wallet and use in applications. So that’s where we’re headed with all of this. And that means we’ll have to do an evolution on the whole identity issuer lifecycle, on-boarding and other things.
Dominic Hobson 14.39: Does that strategy include issuing identities to corporates as well? You’ve mentioned … A company can get one of these?
Kåre Kjelstrøm 14.52: Yes, it absolutely does. The thing about companies is that we have this database called the LEI [Legal Entity Identifier] database, where you keep, you know, information about European companies. That’s a good starting point for on-boarding something like this. And that’s what we’re looking to do, so that you can basically on-board European companies, and you can issue a company identity, which would then be sort of the starting point for bootstrapping issuing identities within the company. So, you can imagine, I mean, you could sign up as the Future of Finance administrator by proving that you have the nationally issued company ID somehow, right? We need to figure out the details. But if you could do that, well then you could start issuing company IDs, or employment IDs to your employees, which can then be used in applications on the blockchain, like proving that you are working for Future of Finance or for Concordium. It might give you some benefits with some partner that we might have, like the 10 per cent discount in the bookstore. Or it could even allow you to vote on certain things within the company, or … you know, you can come up with so many different things that that could be used for. So that’s the direction we’re headed. And we’re planning to do the evolution on all of that in early 2023.
16.12 Are you optimistic that digital IDs will be widely adopted in the near future?
Dominic Hobson 16:12: As far as I’m concerned, it can’t happen fast enough. Every time I’m asked to prove my identity as myself or prove the identity of my company, I ask myself the question, `Why aren’t digital identities wildly popular and being implemented everywhere?’ Do you think do you think that one day they’re going to be? That digital IDs are going to be widely adopted at some point in the future – the near future?
Kåre Kjelstrøm 16.35:Yes, I think that will happen. I think there’s also a lot of resistance in the market. But I think, you know, the way this should be approached – and that’s always how you win it – is to make sure that there are, like, really great reasons for people to do this; they really want to do it, right? So they couldn’t even avoid doing it because the benefits of doing it are just so great. So, for instance, here, in the Fall, we are going to be rolling out a new version of Concordium, where you’ll be able to use your ID – the one that you already have – in decentralised applications. So my application could ask you, you know, `Are you are older than 18?’ before you go ahead and go into my online gambling shop, for instance, right? Or my casino. Because I’m required to do that by law. And using our wallet, you’ll be able to provide us with a knowledge-proof that will actually allow me to determine that you are indeed older than 18, and you hold an English passport; it was verified by this issuer. And if I trust that, then I’ll let you in, right? That opens up for so many use-cases. And if you have, you know, casinos starting to require this because, first of all, there will be the forcing function of the regulation. And then the other side of it is, of course, the convenience of just having to identify yourself once and you can control who you give what information to and you won’t be giving your birthdate and your name and other things that you really don’t want to give to the casino. You’re just giving away, `Yes, I am older than 18, right?’ But not how much older. Just enough to know that you can actually get in. So I think if you can create that, and if you can create sort of an infrastructure where it becomes super-convenient to use the ID, then, you know, the [inaudible] will come out of the bottle. But that’s what we are still pushing for, right? Getting to the point where this actually happens. I don’t know how long it will take but it will happen.
Dominic Hobson 18:21: When you’ve reached that point do you envisage yourselves as Concordium in effect doing that work on behalf of the consumer or the corporate? In other words, owning and controlling and releasing, managing the data? Or do you think that consumers and companies are going to want to own and manage and control their data themselves. Will they use service providers or self-serve?
Kåre Kjelstrøm 18.43: I think the latter, actually. What we are planning to do is we are planning to build an open infrastructure where you can onboard identity issuers that can then issue whatever. And then it all boils down to trust. How can you trust the issuer, right? So we have to get the infrastructure ready for that. And then, using the wallet, you have basically a meta[phor] where you can keep all of these identity attributes at hand and ready to use yourself whenever you feel like it, right? So instead of being in a situation where you have signed up for, let’s say, Facebook login or something like that, where you gave them a lot of information and now the Facebook login button is all over the place, which is great for convenience and probably also great for security – they have two factor authentication, all the good things. But it is not so great for your privacy because Facebook will be collecting information about what sites you’re accessing, and potentially also using tracking cookies. And for Google, it’s probably even worse, right? They also have calendars and Gmail and other things. And so your privacy is really gone. But in this in this space here you can actually have much more control over your privacy, because now you own it. And if you if you fast forward into the future, you can imagine things like electronic health records and other things you can choose selectively to give information away to select practitioners who may not need to know everything about your medical history, but just enough about your blood pressure to treat you for some ailment that you have, right? And that’s where I think things should be headed, like self-sovereign control, decentralized, and completely not controlled by a single entity. And then that’s also what we are aiming for. We’re not aiming to be a central body that does everything.
20.21 How has Concordium overcome the speed and scalability shortcomings of blockchain?
Dominic Hobson 20:21: You are going to make it easy for providers of those services, if you like. Now, you are a Layer 1 blockchain. I wonder how you have dealt with the other issue, which always comes up with blockchain, which is the question of speed and of scalability. I saw that you’re claiming a speed of 12 seconds to finalisation. Speed and scalability are kind of [different] facets of the same thing. And how have you overcome the speed and scalability shortcomings of blockchain?
Kåre Kjelstrøm 20.52: Right, so the finalisation protocol that we have implemented is a result of the collaboration with COBRA at Aarhus University. And it’s basically a finalisation protocol where all finalisation is run by a sub-set of the “bakers” that are the nodes that create the next CCD coin, right? So these are specialised nodes running on the network. In order for the finalisation to work, we actually have a protocol [that] actually requires that we have two thirds of honest “bakers” running. And once we have that, then we can guarantee the finalisation within this time. So I would say that it comes down to the actual protocol that we’ve implemented, which is based on this sub-set of finalisers that [makes] it really easy to come to consensus on what it is they want to finalise and then finalise it.
Dominic Hobson 21.49: This might be a stupid question, to which you’ve already given the answer, but is the solution you have found to the speed and scalability problem actually unique to you? You are obviously working with this other party, but is it just a variant of the sharding and the side-chains, which we read about as solutions? Or is it something completely new?
Kåre Kjelstrøm 22.08:Well, back in 2017, when we came up with this, it was completely new. I would say that these days, others have also implemented fast finalization; we’re not the only ones having this. And then, of course, solutions like side-chains, with roll-ups and other things, have popped up as ways to improve on the speed for some of the other blockchains. Ethereum, for instance, has this. There’s also sharding, that others claim to implement, more or less, right? We are currently actually in … We sort of doing the maintenance on our finalisation layer now – not actually on our finalisation but on our consensus layer – in order to increase the speed of that because, in parallel, while we’ve been researching at the University of Aarhus also have others. There’s a lot of research that’s gone into this particular space. And, of course, we will be leveraging that, right? So we are looking to actually implement a variant of the – it’s called HotStuff – consensus protocol. So a variant of that is what we’re going to augment the existing consensus protocol with. And we are expecting to get a speed of two to three X based on that. But it’s still TBD, exactly what that will yield. And then there’s sharding, which is also really a scalability thing, right? So the consensus protocol update, which is coming in the fall of this year, is sort of the precursor to us doing real sharding. So we are currently designing the actual sharding solution and figuring out how that will work. And I can say to you it’s doing sharding the real way, doing it right and doing it in a way where you can scale out and you can reach out and you can figure out how to distribute your accounts and your smart contracts across multiple small chains. It’s not trivial at all. So the aim for us is to be ready with design around Christmas, and then starting the implementation in Q1 of 2024, with an explicit aim to be able to scale out, right? Of course, scalability is a function of success as well. So, at this time, we can still comfortably live with the 400 TPS [Transactions Per Second] that we guarantee on the chain. We have also seen upwards of 1,000 but 400 is what we guarantee. As we go forward, and we’ve seen more transactions, then obviously this is something that’s necessary. So that’s our strategy right?
Dominic Hobson 24.31:Now, something else which caught my eye about Concordium is that you are open, your are public, you are permissionless. Does that mean that you are happy not to work with institutions? I mean, it sounds like you’re working with quite a few anyway. Or does it mean that you feel that you can provide institutions with an open, public, permissionless network which will meet their concerns?
Kåre Kjelstrøm 24:54: Yes, I think, I mean, the public blockchain is basically a place where you can put information and show to others that, you know, there’s non-repudiation involved in a blockchain, right? Once it’s written, you cannot dispute that this particular set of bytes was as they were at that given time, right? So that’s great for signatures and documents and other things. And so there are certain use-cases where any type of business can actually benefit from using a blockchain, typically in the sort of interface between their customers or their partners where trust is somewhat of an issue, right? So if you can use a blockchain run by a decentralised body of people and companies to prove certain things about a transaction, or who you are, or that something happened, then yes. And that’s, an obvious [benefit], right? If you don’t trust the corporation, as a consumer, for instance, you may trust when the big company writes something on the blockchain for you, that this was actually what happened. Or that, you know, you transfer the deed or something between x y, z and it actually sits down on the blockchain now for everyone to see. So I believe, yes, I think that’s … It doesn’t preclude us from working with institutions. On the contrary, I think there are lots of use-cases where this can be used.
Dominic Hobson 26:19: You’ve also chosen a Proof of Stake model. Now, Proof of Stake models have acquired a reputation for having what might be called governance problems. How are you managing those in your Proof of Stake model?
Kåre Kjelstrøm 26.33: It’s a very good question. And another one where this whole aspect of running a decentralised blockchain comes into play. And we’ve touched on it right? Like everything you do when you run a public, permissionless blockchain, essentially has to be decentralised. And I mean, you start out … Whenever you start something, you always start out with a centralised body, some people coming together working on something [like] publishing, you make a company, and then you start it up. And the same thing is true for Concordium. We have a blockchain right now, which is decentralised. But there are some centralised aspects that we are looking to decentralise. And one of them is the governance aspect. Right now, there is a governance committee that meets every so often to decide on changes to tokenomics, on changes to key operational aspects of the blockchain, and also potentially what is sort of next, right? We’re looking to divide those two things. So we’re looking to have an organisation that builds the blockchain and evolves but which is actually being operated, or paid by the Concordium blockchain, which is then governed by a de-centrally elected set of people or by way of direct democracy. Now figuring out how to do this is almost like creating a new country and deciding what governance you want there, and what type of government do you want, right? And so we’ve partnered with the University of ETH in Zurich. I mentioned I have scientists sitting in Zurich, and they are collaborating closely with ETH as well, where we also have a great set of scientists, both within tokenomics, within finance, but also within crypto, that can help us out with this. Specifically, we’re working with Professor Gersbach and his group at the university to figure out what should, basically, the government’s model be. Should we base it on who has an account already? On who has a certain amount of CCD staked on the blockchain, or no? What are the functions that we want? Do we want direct democracy for some decisions? How do you even know what to vote for, right, as a person who can vote? And then how do we make sure that, you know, we don’t create an infrastructure that can be used negatively for a hostile takeover, which can then, you know, bring down the chain potentially, right? There’s like so many things to figure out. But the intent is to have a relatively fast working group that will have the essentials of this worked out in early 2024, so that we can start implementing a DAO [Decentralised Autonomous Organisation] and start looking at doing elections in 2024 on small things, right? And then potentially elect a person from the outside in as one of the first things, and then start, you know, making votes about smaller things and then as we progress over the years, we will gradually decentralise Concordium’s governance body even more. But we are very wary that this has to happen slowly and you cannot force it because you might easily make a mistake, right, that brings down everything. So it’s super dangerous and not trivial.
Dominic Hobson 29.38:I can see that your DAO – your decentralised autonomous organisation – is still at least 18 months away. But I don’t know whether your friends at universities have been able to decide how a DAO fits inside existing laws in either Denmark or Switzerland. It’s a live discussion here in the UK, you know. Is it a partnership? Is it a company? Is it a co-operative? What is it? In the US they seem to – in certain jurisdictions – have decided it’s a co-operative. But, looking forward to when you have this in place in, say, 2024, do you think that DAOs are going to mature to become a genuine alternative to the corporate structures we have today – by which I mean limited liability companies and limited or unlimited liability partnerships? Those are the two basic structures plus, I guess, a cooperative. Do you think they provide a long-term alternative?
Kåre Kjelstrøm 30.31: I think it’s a very interesting question. And it also, I mean, the fact that we’ve been talking about it kind of shows the typical scenario that technology is pushing the boundaries of law and regulation. A DAO is, I guess, conceptually similar to an LLC [Limited Liability Company]. There’s no leading member of the organisation that has majority rule. So in this case, you can set up a DAO that does the same thing. In a DAO, the whole thing is sort of modelled on a decentralised cryptocurrency and it operates without any human intervention, so it’s all just a set of smart contracts that clicks on the chain. And by that you can do all of your governance. Now, I think we can all agree that we could implement this, and we can test the smart contracts, make them safe and secure and convince ourselves that this would actually work. And we can even get it up and running. But getting from there to getting countries and the legislation in place for this to be recognised as an LLC or something like that, I think there’s a ways to go there. And I think that we have to approach this piecemeal and there has to be legislation worked slowly, just like we should with the governance, right, and figuring it out to do it right. And I think it’s not actually necessarily a bad thing it works slowly because, again, it’s a tricky thing to get right. We’ve spent a lot of time outside of the blockchain world over the many, many years that companies have existed to get this structure right, and then getting it into a new world is something that has to be done in the right way. But I think yes, potentially, yes, it has it has the potential to become such a thing in the future.
Dominic Hobson 32:10: Talking of cryptocurrency, Concordium has also issued a native coin of its own on to I think six cryptocurrency exchanges. Now is that coin used primarily to pay the stakeholders for the work that they’re doing in your Proof of Work model? Or does it have a wider role to play, such as settling transactions on the blockchain network?
Kåre Kjelstrøm 32.33: The interesting thing about a decentralised blockchain, a public, permissionless blockchain like ours, is that in order to incentivise anyone to run the nodes … I mean, what would incentivise you to download my software and, you know, turn on your computer, install it and have your computer running 24/7/365 and consuming power? Well, not much, unless you did it for altruistic reasons, or because you were incentivised otherwise, right? So in order for that to happen you have to have a currency on the blockchain. So that’s really the basics of why there’s a currency. And now in a Proof of Stake blockchain, we have the miners or the bakers as they’re called in Concordium, which are these specialised nodes that have a chance every so often to create a new CCD. And that new CCD will then be part of the reward, and they will help do the infrastructure of the blockchain and do the consensus and do the finalisation to make sure that all the transactions are written and in return they get the CCD. So that’s really the basics and you can have the blockchain running and clicking there. If it weren’t for the fact that it every time you do a transaction on the blockchain, there’s a small fee involved as well, right? If you do Ethereum, you are looking at gas fees in order to do a transfer between accounts or if you want to do anything. So for you to do anything on it, you’ll have to go and get some CCDs basically, in order to do transactions. So, I would say that the CCD is the foundation for kicking off a whole, like, wide array of applications. The whole decentralised finance space starts with this and the fact that there is a coin. There are things, basic things like you now have your CCDs on the chain and you want to counter that there is an inflation on the chain. We have an inflation, obviously, because every time a small amount of time has passed, a new CCD is minted [and] that means all of the other CCDs just slowly decrease in value for every single CCD that is minted. That inflation you want to counter, how do you do that? Well, if you could put them somewhere where you could make a buck, well, wouldn’t that be great? And that’s why we have staking, right? So, as a person, you can maybe have, let’s say you have 1,000 CCDs in your wallet, if you just let them sit there, they depreciate over time. But if you decide to help participate in the blockchain, by putting your stake within a baker, you can do that on Concordium – basically, it’s taken directly from your wallet – you can make an interest, right? And then you can make … you can basically improve on your wealth and counter the inflation. And of course, that’s just the primitives, right? It opens up for lending protocols, decentralised lending. I can lend you money, and you can borrow from me without an intermediary, such as a bank. And if you have, like the ID layer that we have, I might be inclined to also want to know your name and and learn more about you, right? And you can even prove to me that you might even in the future be able to prove your address as well, if I want that, right? So that way we can establish a more firm foundation of trust, and I can now start doing transactions with you directly without having a bank. And so all of that, again, opens up for things like gaming and Metaverse, where money and currencies is also going to be a thing, or is already a thing. So yeah, I’d say that there’s so much to say about it, right? But it’s sort of the foundation of a Proof of Stake or even any blockchain – also Proof of Work blockchains that you have this currency. And, as a side effect, you can build all of this interesting stuff.
Dominic Hobson 36:12: Now a question about bridges, Concordium has built bridges to other blockchain networks, for the obvious reason that inter-operating with other networks is good for every network. It creates liquidity, and so on. But those bridges have acquired a reputation for creating points of vulnerability. How do you make sure the bridges you’re building are secure?
Kåre Kjelstrøm 36.31: It’s an excellent question. So, yes, we have a bridge, which is going to be launched very, very soon. We actually don’t have a bridge at this time. But we will be launching a bridge called Cornucopia – like a horn of plenty, because we’re hoping that a lot of coins will be exchanged over that bridge. The launch date is not set yet, but it’s likely within a few weeks. It’s already built and tested. In order to verify that the bridge actually works, we are doing quite a few things. What everyone does is they work with some external auditor that has a long checklist of things to look for in the smart contracts. And in particular, the smart contracts on the two chains that the bridge operates between have to be verified intensely. Is that enough to make sure that you don’t have problems? It’s not, right? There are many examples of bridges that have been compromised or someone was able to mint a rap version of a coin without actually putting the real coin in on the other side. And then, after that, they could sell the whole thing back to the bridge and deplete the bridge for Ether or something like that, right, and then bring the whole thing down. So we do that. And we also work again, with the University of Arhus. There is a research group over there that does exactly this. They’re working on smart contracts and formal verification of smart contracts. And we’re currently formalising how we will be utilising what they have already and how we can evolve formal analysis of smart contract code, in order to make sure that what’s going on here is actually really the real deal, right, and that there are as few bots as possible. I will not sit here and guarantee that that means that in the future there will be no bots in our smart contracts, because there are always bots in code. And there might also in the future – fingers crossed that it won’t happen – but there might, of course, also be a situation where we have smart contracts that are compromised. I’m hoping we can avoid that. But I also know from experience that there are bots in software. These are basically the ways, right? So, manual review, analysis of the code using tools, and then formal analysis. And as this is a research field, which is fairly young still, I’m expecting lots of things to happen there. As you narrow down the scope of what you can express in the smart contract to just the right things, you can also create more formal verification methods that can check for all kinds of strange vulnerabilities in the code.
Dominic Hobson 39:01: Now I probably should have asked you this earlier, but it’s a question about zero knowledge proofs which, crudely speaking, I understand to be using hashes to prove authenticity without sharing confidential information. What uses – what value – have you found for zero knowledge proofs in your work?
Kåre Kjelstrøm 39:18: Zero knowledge proof is actually used for quite a few things already in Concordium. So, during the ID registration, we use it to show that the user knows the relevant secret values and found the data there correctly. When you open an account, we use it to show that the user has a valid ID and that is correctly attached to the account. We are also currently rolling out – or it’s going to be rolled out in the Fall here – the ability to recover an ID from an identity provider. There, you have to re-authenticate with the identity provider and prove that you’re you and then you can recover your ID. And then we have a feature called “shielded balances,” where you can basically hide the amount of CCDs that were transferred in a transaction so that it’s not shown on-chain for everyone’s eyes, but only between the two parties that were interacting in the transaction. This is also using zero knowledge proofs. And then when we release the new version of the ID in the ID 2.0 in the Fall, it will be used also for me to prove to you that I’m older than 18, or from Denmark or something. And now we can start building applications around it, which is a very cool application.
Concordium has identified a wide range of primary uses-cases for the Concordium Layer 1 blockchain, including tokenisation, automated market making in DeFi, DEXs borrowing and lending apps, digital IDs, supply chain traceability and settlement. insurance contracts and GameFi. Why did you choose them and what ties them together?
Dominic Hobson 40:36: Use cases – you’ve touched on these once or twice in this conversation. When I go to your website, I see a list of use-cases you’ve identified as your primary targets, if you like, for this Concordium blockchain and I’ll list them for you. They’re tokenisation, automated market making and DeFi protocols, decentralised exchanges, borrowing, lending, and so on. The third one is, less surprisingly, digital ID. There’s supply chain traceability and settlement. There’s insurance contracts, there’s also GameFi. Now, I don’t know whether you can give a single separate reason for why each of those has been identified as a use-case, but what does that list tell us about where you see the opportunities lying for your particular blockchain?
Kåre Kjelstrøm 41.32: I wish we had published our new website. We have a new website. It is ready to be published and it will be out again in a few weeks. On that website you will be able to see a slightly different message. But everything you mentioned here is still relevant. And so we are definitely also looking at all of these things. I think at the foundation of Concordium is the ideas I mentioned, and we look through all of these other things through that lens. Tokenisation, of course, is what has been sort of a very, very successful use-case for many blockchains over the past couple of years, two-three years. We’ve seen those coins and bought [Bored] Apes and other things pop up that have gained massive value and that people have been exchanging. And now we’re looking at tokenising the real world, right? Tokenising cars and car ownership and those things. And, of course, we want to be able to make it super-easy to build these kinds of applications on Concordium. Like make it easy to make a non-fungible token [NFT] about something. To use Concordium to ping IoT [Internet of Things] devices in the real world. And, again, IT infrastructure can be used to identify devices as well, which is the thing we’re looking at as well. How can we tie that in with real world devices that you then can prove ownership of and other things. And then, of course, other things in terms of fungible tokens. You can provide a new kind of currency on the chain. Or even these non-transferable, non-fungible tokens, also known as Soulbound Tokens [SBTs], which is what we are going to be using to control the lifecycle of identity attributes on our chain, which should also be just, like, super-easy. So for tokenisation it’s really about lowering the bar, making it super-easy to build these things, and then leveraging our ID to be used where it makes sense, right? DeFi? I think we touched on that already. And I think for DeFi some of the problems that have been tormenting the space – namely, that it has been used for money laundering or, you know, illegal activities – we are hoping that the infrastructure that we bring to the table will actually help counter some of these problems. And then, of course, also, like I said, lending, for instance, right, there are situations where you want to reveal more about yourself, in order for the other party to gain trust, that this is actually a person I can rely on KYC protocols. We’re not doing KYC, by the way, we’re doing identity verification. Full-fledged KYC would also require you do checks against various lists and other things. But having a KYC process where you can actually take your KYC information and use it in different settings. Today, most countries require financial institutions that they do their own KYC. So every time you change a bank, you have to do your KYC again, which of course is annoying for the consumer. But imagine if you could get to a point where you can actually do the KYC once and then, because it’s been through a process that everyone likes or feels feel good about, then you might be able to re-use that in certain settings and reveal certain aspects from that KYC process, which again, yield some identity attributes. And I think, you know, there are just so many things that we can do in DeFi where the ID is relevant. So that’s a key thing. And of course, the ID, which is the main thing we are pumping out right now as a feature, like I said, is the thing. And then you mentioned supply chain traceability and settlement. The moniker there is ESG [Environmental, Social, and Governance]. And it’s really about both CO2 tracking, but also being environmentally responsible and socially responsible, and then tracking supply chain, right? So the whole thing about tracking carbon dioxide emissions, and proving to the world that you are actually environmentally friendly, in your production of things, is exactly where a blockchain fits very well, right? Because here you can now publish that at this point, you know, I moved my goods by truck from A to B; this is exactly how much gas was used and this and that; therefore, the CO2 emission was such and such. And therefore the [inaudible] that you’re buying has had the CO2 footprint. So I think those things, getting into that vertical, is a space where things are happening, and that’s why we’re looking to do that. In terms of insurance, and insurance contracts, the blockchain has the ability to help automate claims functions, like by verifying coverage between companies, and reinsurance. But it can also automate payments between parties for claims and can help lower administrative costs for insurance companies. So we think insurance is another one. It’s also a place where the ID plays a role and potentially also KYC plays a role, right? If you sign up for a new insurance provider, you also typically go through a long process of giving a lot of information about yourself. And imagine you could do that once. You have self-sovereign control of it. You give exactly what they want, or you choose not to give it because you don’t want to give it away and you go to another provider. And of course, games, Metaverse, that whole space there, is huge, right? You’ll see that some of the … We are currently collaborating with quite a few games studios on building on Concordium. And this is both for things like in-game tokens, like an economy within the games. It’s like assets, where you have an asset within a game. And imagine if you could actually sell that – you can take it outside of the game, or you could re-use it in another game, published maybe by the same provider. Now, all of a sudden, the asset takes on a life of its own. It becomes an NFT. You can even showcase it on some websites or your social media. And then things like peace of mind for parents, right? The fact that this game is recommended for people older than 16, and your 11-year-old wants to play it, you just turn on `No sorry,’ right? Not happening. This is a space where, you know, there’s a lot of opportunity there. Also things like chat rooms is a feature that’s there in many on-line games and or even just voice chat, right? And then you know, you’re playing with this avatar and the avatar becomes your friend and you feel like you want to meet up in the real world. But how do you know that this person is actually this 15-year-old girl that they claimed they are? How do you know, right? And so, again, the ID can come into play in order to prove that this is indeed true, and that you should feel more comfortable about this person that you’re talking to.
Dominic Hobson 48:17: It’s interesting, and obviously very significant, how central that digital identity piece is to so many of the use-cases that you’ve just talked about.
Kåre Kjelstrøm 48:26: Yeah, I think also that’s what we identified, right? If once you start looking at it, for all of these things, there is like, `This is a key aspect that can help prove trust and confidence between the parties interacting.’ And the fact that, you know, we allow you to selectively give away some information – not everything, maybe just the fact that you’re from England might be enough for me, right? Or the fact that you are older than 21 might be enough for me, or the fact that you’re not from North Korea or something – [means there is ] something that could be enough to establish the element of trust that I need in order to do business with you. Or that you can give away your name and I can actually know that that that your name is indeed Dominic, right?
49.10 How many clients is Concordium working with and how many have you signed up?
Dominic Hobson 49:10: Now, just before I let you go, some questions about where you’ve got to. How many clients you working with? How many of you signed up?
Kåre Kjelstrøm 49:17: That’s a good question. I would say that, you know, we have a couple of applications running on Concordium now. The Concordium mainnet was launched a year ago and we started selling the CCD in February. So we’re now building out the infrastructure to really make distributed applications that will come now, right? So we’ve sort of been going slowly also because of the science thing in it. But we do have a couple of things. There is a naming service running on Concordium. There is an NFT marketplace, SpaceSeven, running on Concordium. And a few other applications. We have quite a few that are coming, so the pipeline is bigger. We have a grants programme that is being used to fuel all kinds of applications. I mentioned For Games that we have signed up with. We have an energy provider. We do already have a Japanese party that’s using Concordium for e-voting. We just started the collaboration on a tender with the government of Greenland, which is an interesting one. The government of Greenland is looking to do e-voting, and so over the next three years we’ll be evolving how can you do e voting on Concordium, in collaboration with the University of Aarhus. And yeah, those are those are some of the bigger ones that we have. But I mean, we don’t have in the 1,000s at this time. But of course, as soon as the ID comes out, I mean, I’m sure it will explode, Dominic.
50.43 Who do you feel Concordium is competing with?
Dominic Hobson 50.43: And in terms of who you feel you’re competing with?
Kåre Kjelstrøm 50.49: We are obviously competing with all of the other blockchains. But, in reality, I think if you look at what the space is like, you can view it from two angles. One is that you want to be the next Ethereum killer. Or do you do want to position yourself within an ecosystem of blockchains or even within an ecosystem of IT, Open IT running on the internet? And we are looking to do the latter. We don’t think that Concordium is going to blow Ethereum away. On the other hand, we think that there’s ample room for a lot of chains out there with specialised capabilities. And we’re looking to actually take the ID to a place where it can be used in white label solutions, and where you can use it as an infrastructure, if you want to attach an ID to some of the stuff you already have. So imagine you have a decentralised exchange or something running on Ethereum, and it uses Ether and that sort of thing, but now regulators are knocking on your door and, you know, require you to have some kind of way to identify your customers with safety. Wouldn’t it be great if you could just plug in a little thing into your wallet, and then boom, you could go through the IDV [identity verification] process. And all of a sudden you can start transacting with identity and that identity was actually all identity that we provided as an infrastructure. So that’s one of the future directions that we’re looking to go. And I think that will be great because then, you know, we are positioned as an identity solution to anyone else. But we also are a blockchain on which you can already build these applications. You don’t have to build it anywhere else. You can just build on us and get both right. So we’re not looking to blow everyone out of the water. We are looking to collaborate and actually provide a service for everyone.
Dominic Hobson 52:34: I wonder what your investors are looking for. We talked at the outset about Lars Seier Christensen. He’s obviously a shrewd entrepreneur. He sees transformative potential in this technology. But he’s not the only investor. The company’s raised US$30 million, according to Crunchbase, from other investors as well. What do you think they saw in what Concordium was doing that they didn’t see elsewhere?
Kåre Kjelstrøm 52:59: Yeah, we actually raised US$50 million in a private placement.
Dominic Hobson 53:02: Right okay. The information is out of date, or I am anyway
Kåre Kjelstrøm 53:06: But I mean that is even better, right? So, I think the investors that we have are here for the long haul. They believe that the world of blockchain will need the infrastructure that we are providing. They also see the same picture that we do. They realise that, eventually, regulation is going to be coming down hard on this whole space. And if you don’t have an answer to how do you do more secure and safe transactions in terms of who you’re dealing with on the blockchain, then you’re going to be in a situation where you might not be able to continue to operate. So they’re betting on this horse because they also feel that building an infrastructure around the ID is going to make it [inaudible] in the future. So yeah, I mean, I’m in dialogue with many of our investors at this time and they’re happy with the direction we’re taking and are backing this up 100 per cent. So I feel like we are definitely headed in the right direction.
54.10 What will success look like for Concordium?
Dominic Hobson 54:10: Well, this is my last question. How will you know when you’ve succeeded? Do you have a picture in your mind of what success actually looks like and, when you get there, you’ll be able to say. `We have succeeded’?
Kåre Kjelstrøm 54:22: Success to me looks like, you know, games where parents don’t have to worry about what their kids are dealing with; a Metaverse where the same thing is true; finance where old money can come in with peace of mind and doesn’t feel threatened about dealing with, like, scamsters or people that they can’t control. And then, of course, it looks like lots of transactions going on on the Concordium blockchain, lots of activity, the chain growing. But first and foremost, customer success and customer happiness. I mean, if you take care of your customer, your business will take care of itself.
Dominic Hobson 55:03: Kare Kallstrom, thanks very much for taking the time to talk to Future of Finance.
Kåre Kjelstrøm 55.07: Thank you, Dominic.