Blockchain. Is it the most revolutionary technology in value exchange ever or just the latest fintech buzzword enjoying its peak on the hype cycle? Or both? These young techniques are undergoing swift evolution, going from bitcoin and money transfer into new use cases such as identity management.
There’s still truth in the cartoon’s joke that, online, no one knows you’re a dog. The challenge goes beyond discerning hacker activity from the permitted. It’s also about the release of just the data necessary to satisfy the needs of both parties in a transaction – and no more. How much better that would be than sharing a full suite of personally identifiable information when simply asking the question “are you over 21?”
Take a listen to my discussion with Matthew Commons, CEO, of Cambridge Blockchain on how his company’s blockchain-based approach can be used to address one of the internet’s remaining fundamental concerns. You’ll also learn about the state of play in this new stage of permissioned blockchains.
Payments on Fire 31
George: Welcome back to Payments on Fire, I’m George Peabody with Glenbrook Partners, and today we’re going to talk about the evolution of blockchain technology. At Glenbrook we actually kind of look at blockchain across three different iterations. Blockchain 1.0 being, of course, bitcoin and the flock of cryptocurrencies that it inspired. Blockchain 2.0 is actually where we see blockchain technology, and frequently, it is in fact bitcoin ledger that’s enabling it. It’s a movement of money, where Blockchain 2.0 really fits into financial services, clearing and settlement, international payments, B2B, remittances, trade finance. Where there’s a great deal of activity right now is in what we’re calling Blockchain 3.0. This is where blockchain ledgers are being used, for multiple applications, to track the flow of assets of all kinds across a ledger, to enable smart contracts and smart property. Of course there are applications that are being talked about in the world of the internet of things. Some of those are quite broad scale and ambitious, and there’s some really very intriguing and perhaps potentially broad scale, but they can enter a tactical point of view, and that’s another use case which is around identity management. It’s my pleasure today to have Matthew Commons, who is CEO of Cambridge Blockchain, who is working in that space with us today. Hi, Matthew. Welcome; glad you’re here.
Matthew: Hi there, George. Thanks for having me.
George: So, Matthew, we’ve talked before. Why don’t you tell me a little about Cambridge Blockchain and what inspired you to create your business.
Matthew: Sure. Cambridge Blockchain creates an identity management platform for blockchain technology, and it really solves one of the key issues for getting these – some Blockchain 2.0, but really it’s Blockchain 3.0 – applications out of the labs and into a production environment. The key issue that we’re really solving is how do financial institutions, within a blockchain context, do the appropriate identity checks that they have to do under Know Your Customer rules, but still maintain customer privacy if all that information is on some type of a shared ledger system.
George: Ok, so the ledger you’re working on is around identity management itself, not about moving money from one account to another.
Matthew: Our technology is really an enabling technology for all different types of things that you need to be able to do on a blockchain. The challenge of trying to get strong identity and also strong privacy protection in the same blockchain system is a fairly common one. You’ll see groups like Ripple who have been fined over this, groups like R3 are holding workshops on it, banks are blogging about it, and so how to get identity and privacy right in blockchain systems is really the key area that we’re focusing on. It’s an issue that’s sort of asset agnostic, it doesn’t necessarily matter whether you’re transferring payments on the blockchain or whether you’re transferring some type of smart property, or is it just a ledger, or is it just a smart contract – you have to get identity right. This is a key issue for banks. Banks have spent over 2.5 billion dollars on compliance last year, and a big part of that was around identity management. Certainly, when they look at going into a new technology like blockchain, this is something they absolutely have to get right.
George: Is this an identity management system that an individual financial institution would operate?
Matthew: With blockchains as a category, there are pretty strong network effects, so it is something you can operate within an individual institution. It’s an internal ledger, so we are working with one bank, for example, that is looking at their internal ledger to have different parts of their bank communicate and be working from the same sense of data, so they can have much more straight through processing of types of transactions that involve different parts of the bank. So you can do it that way.
George: That would have a single authoritative ledger or blockchain for identity management across all the touch points that that financial institution’s got out there.
Matthew: Yes. What we’re doing, it’s really a piece of the puzzle. It’s a very critical component of the blockchain stack that you have to get right, and then you can then use that to do all types of other blockchain transactions. Our platform can plug-in to other types of blockchain systems; it could be a private version of Ethereum, it could be some type of transaction only ledger, it could be for transferring bonds, it could be for transferring customer data, it could be for maritime finance. Ultimately, all of these use cases for blockchain have this issue around identity and privacy, and that’s the issue that we focus on.
George: Got it. So, you bring a blockchain technology that addresses the identity problem.
Matthew: That’s correct. It plugs into a broader blockchain technology that is going to be much more customized, more for the specific use case. If we’re plugging into something for maritime finance, that type of system is going to be set up in a different way than something that is a group of banks trading corporate bonds. It’s the same underlying issues that you have.
George: A fundamental identity issue is to access each of those individual blockchains built for specific use cases.
Matthew: That’s correct.
George: Got it. Talk to me about the network effect. I get that the more people who are participating in a blockchain based identity system, the more robust it becomes. Who become the operators? What are the incentives to participate?
Matthew: It does depend a little bit on the use base. You know, blockchain as a whole, they have strong network effects. You’re going to feel sometimes like you’re trying to sell the first fax machine. A fax machine by itself is completely useless, you got to sell them in pairs, and I think that’s no different for any of these types of blockchain systems.
George: Great analogy.
Matthew: In each different type of market there are different dynamics in terms of how many participants do you need to have in a system to get a critical mass, not from a technological point of view, but really from an economic point of view. Do you need to shift the whole market over to this blockchain system in order to get started, or do you have some sub-set of that market that is on there. I think what is going to make different blockchain use cases successes or failures is their success from an economic perspective – can you get people together, get people abiding by the same regulations and really working on some kind of a common system. What that typically would be in applications that we’re looking at – and we’re looking more at conventional financial institutions, large banks, insurance companies, things like that – there would be basically a group of banks or brokerages that are getting together to set up a blockchain system for a certain type of asset transaction. It could be bonds, it could be commodities, and it could be something around trade finance. Ultimately, our technology platform is asset agnostic, but there are these groups of financial institutions setting up these networks and the specifics of how they set it up will dictate then the best technology choice. I that I think is being commonly used out there right now are basically private versions of Ethereum, which is the smart contract platform. The R3 groups of banks announced that they had done a test using that private version of Ethereum, a lot of what we’re doing ourselves. We also use things on private or permission, probably the more appropriate term is Ethereum. This is not on the public version of the Ethereum smart contract platform, but you basically have a private port, and you have a group of known institutions that are transacting amongst themselves. So you set it up in a little bit of a different way, you don’t necessarily have the same type of mining based system that you have in the public version. We’re using something called proof of authority instead of something called proof of work. Proof of authority is basically you’re dealing with white list of no’s, and you have a process for bringing your no’s into the system. The security of it really comes from how to maintain that white list a certain way so that you are robust to malicious attacks.
George: You have a community of interests interested in protected it’s joint interest. I can understand how that method would be appropriate. You don’t need an open platform.
Matthew: That’s correct. Most of what I think you’ll see the large financial institutions looking at are not open platforms.
George: Of course not.
Matthew: I think it’s important to note that we do think that there will be common protocol for how the blockchain platforms communicate with each other. The W3C, the World Wide Web Consortium, is doing a lot of work on that right now. Our chief technology officer, Alex Oberhauser, is quite involved in that. We see that it’s not going to be a world of everything is on the bitcoin blockchain, or Ripple is the main thing for payments, there’ll be different blockchains set up for different use cases, but there will be these emerging common standards of how they communicate. So if you loan something on one system, you can still prove that on another system, and you can have asset transfers between different networks of blockchains.
George: Is this a sidechain technology, or is it an interface challenge?
Matthew: Yeah, I think that’s a better way to do it, you can’t set things up as sidechains. I think most of the time when people talk about sidechains, they’re talking about it relative to the bitcoin blockchain, but this comes to pegging chains, you can do that but you can also really prove ownership or you can prove that you can control the cryptographic keys or new assets on one chain, you also control the cryptographic keys on another chain. There is this emerging language of how these cryptographic proofs will go across chains.
George: How far along is that?
Matthew: There is some degree of agreements on some of the basics. There are a few different approaches, I know Ripple has one approach – their interledger project, the W3C, this group that Alex is working with has a slightly different hack, and I think there’s ultimately some different platforms out there. Ethereum is probably the one that is most well used for this right now. There’s also others. IBM and the Linux Foundation are doing a lot of good work on taking some of IBM’s work and open sourcing that. We have to really see how this evolves over time, but ultimately, what we’re focusing on as a company is this key issue of managing identity and privacy, and that is really an issue that’s common to all blockchain platforms in all asset classes. Where we look to implement, that is really going to depend on what really gets traction in the market – so if it’s treasury repos, then we’ll be focusing on treasury repos, if it’s corporate bonds or some syndicated loans or something like that, then we’ll do that. We look to really be the experts on identity and privacy in the blockchain context and then partner with others that really bring that demean expertise. Clearly there’s many, many applications for this and for a small company like ourselves, we certainly can’t be experts in all these different areas.
George: As I recall from our conversation months ago, that one of the characteristics of your version of the blockchain with respect to identity is that you’ll capture a credential, build a hash of that credential (a pdf of a passport for example), and write that to the blockchain. Then you’ve got a layer that sits above that which mediates access to that record. Is that right?
Matthew: Yeah, the key thing in terms of this is we don’t actually do identity validation ourselves. In most of the models that we’ve been doing, it is the banks or the brokerages that do that. The reason for that is not necessarily a technological reason, but it’s really a regulatory reason. Under the Banking Secrecy Act, the banks have the requirement to do that, and they have to do that themselves. If I have an account at CitiBank, and I want to go buy a product from Santander, Santander cannot rely on CitiBank’s attestation of what I’m doing. The banks have to do it, so if you’re trading on this blockchain platform, whatever the asset class is, the model that we have is basically going to be going through a bank, like you do for trading conventional securities. That bank does the identity check the same way they do now – they look at your passport, utility bills, pull the data file about you, and so forth. What’s different in this system, is we have them sign that information and store a cryptographic proof of the information on a blockchain system.
George: That is specific to the financial institution?
Matthew: Not to that financial institution, that is shared throughout the blockchain network. So if it’s a group of banks trading bonds, having their clients trade corporate bonds between themselves, this identity blockchain would be shared in the same way the blockchain that would represent ownership bonds would be shared.
George: Within this permissioned community, you would have a blockchain for the asset that’s being traded and another one around the identity.
Matthew: That’s correct. There’s a few different ways to set that up, but that’s sort of the basic configuration that we have. What’s important about this is that you have to really think about what are you sharing, even within this permissioned community, you really don’t want to share everything with everybody. So, in fact, you don’t put anything, like a copy of your passport or even your name in something that is shared, but you do have a proof of that. So, if I want to sell you a bond, you may not want to actually tell me who you are, because you may have privacy concerns, you may be worried about me front-running your trades, and any number of things you wouldn’t want to tell him who exactly you are. There may be certain things I have to verify about you. I need to know that some bank has done a KYC check on you, I may need to know your citizenship, may need to know your credit quality, maybe there are some firms in the system that I want to trade with that are on a white list, there may be some firms that I don’t want to trade with that maybe have poor credit quality, and I will want you to prove to me that you meet the criteria that I have to trade. What our system allows you to do is to share proofs to validate that in fact your bank has signed off that you are a US citizen, that you are a member of this white list of firms, that your credit rating is this, and you can selectively release just the pieces of information that you need to complete the trade. You don’t actually need to tell me who you are, you don’t actually have to tell me what firm you’re with, but you can validate these things, and that’s really the key. It’s not about validating the identity to your counterpart, it’s about validating identity attributes. You want to validate just the minimum amount of attributes that you absolutely need to do to complete the trade. You want to share that information only with the counterparts that absolutely need to know that, and everything else you keep private. That’s the key to making this work.
George: So that’s part of your value add, it’s not just around the blockchain, but it’s about the mediation of access to the data that’s required for the transaction, but more.
Matthew: That’s right. That information is stored off-chain, so for a copy of your passport, if you wanted to share a full copy of your passport, that wouldn’t be on the blockchain, but it would be in a virtual container, something called personal data service. It was actually a concept, really innovative, but a group at the MIT Media Lab, which is right across the street from us here.
George: Is this the mustard seed group?
Matthew: Yes, the open mustard seed project. They’ve done a lot of work on this over a number of years. John Kiplinger really being the leader of that project and one of the leading thinkers of the world on this. He’s very active in the world economic forum and others around it. We really take those concepts, like the open mustard seed that the MIT Media Lab has innovated, and really bring them into a blockchain construct to solve specific problems for large financial institutions that are looking to set up these permissioned blockchain networks between themselves.
George: This has been very interesting. Let’s go up to 50,000 feet. How do you think these financial institutions are doing on their blockchain projects?
Matthew: Clearly there’s a lot of interest and excitement in the space. We see a lot of funding flowing in there, whether its digital asset holdings, a recent fund raised of over a 100 million dollar valuation from a company there, the R3 groups getting together. We’re just a little bit pleasantly surprised that you could get 41 or 42 banks all together in one consortium, knowing how typically slow moving banks can be sometimes and how difficult it can be to build consensus. Just getting everyone around the table is a big step. We’re encouraged by that. You’re seeing different use cases come out, and I think people are really trying to figure out where this is going to gain traction first. There’s a few different philosophies around that. Some think that it should be really the biggest markets and the most liquid markets, if you can do this there’s going to be cumulatively the most benefit, so you should go after those. Other ones are saying you should go after the less liquid markets, the ones that take 60 days or more to issue, and those types of things would be proportionately a much bigger percentage savings in time and cost, but maybe a bit of a smaller network to start with, so I think you’ll see this. There’s also different approaches on do you issue assets on the blockchain to begin with and you have that corporate bond issued as a blockchain bond or do you have it issued a conventional way, and someone like a custodian bank holds it and issues blockchain tokens that represent a perfected ownership interest in the underlying security that’s held in a more conventional account. There are different approaches here, and I think 2016 is the year you’re going to see a lot of experimentation. I think you’re going to hear about a lot more prototypes, a lot more proofs of concept in different asset classes, and I think it’s really going to be a learning experience. I think it’s still an open question about which assets are really going to take off here. I think one thing that’s clear is that there’s huge amounts of potential here and people can see, they can really visualize a world in which we can get to in not too large a number of years, where you actually have a majority of financial assets that are represented on the blockchain, settlement happens almost instantaneously, compliance reporting is something that is more or less automated, risk reporting is something that is automated, and these systems just function a lot more smoothly with a lot less human intervention than you have right now. They key is, how do you start from where we are today and then get to that feature, and that’s what we are looking at right now.
George: How do you get there? My takeaway is that while, to your point, a great deal of technical experimentation has to happen, it’s really going to be about the business relationships and the economics and the most appropriate use cases to get this whole train moving.
Matthew: That’s right, and I think that’s really why we have chosen to focus on the identity piece. We don’t really know exactly which use cases are absolutely going to take off, but we know that doing identity and privacy really well in a blockchain context is one of the most critical issues to getting these projects out of a lab and into a production environment. So that’s where we’re focused and that’s where we offer solutions that are better than anyone else.
George: That’s great. I think you’re very wise to be focusing obviously where there are real applications in the short term, because I believe we all know that in the long term privacy is going to raise the level of importance and focus from a regulatory point of view is going to increase over the next 5 years. The activity we’re seeing in the EU already with respect to the care of citizen data is already getting to be an issue. We’re going to need some systemic solutions. Glad to see you’re starting in appropriate sized use cases.
Matthew: That’s great, and we were really thrilled actually just last night that Santander gave a little vote of confidence in us. We were the winners of their distributed ledger challenge, so nearly 150 companies and judging panels including folks like CEO’s of Ripple and TradeBlock and Blockchain.com and a number of prominent BC’s and things. Hopefully that’s something that’s going to help the market see that this is a real technology out, solving a very important issue to enable Blockchain 2.0 and Blockchain 3.0 transactions.
George: Great. Well, Matthew, thank you very much. It was a great conversation; I learned a lot.
Matthew: Excellent. Thank you very much, George; I appreciate it. Bye.