I’ve been asked to explain what a blockchain is, and we’ll go over that material in a little more detail than usual so that we have a firm platform for understanding for the rest of the day and to catch anybody up who is newly interested in this field, and then we’ll talk a little bit about what the implications are for largescale systems on the blockchain, particularly as they effect the areas that government are interested in.

The first thing is that not all software is regulated or needs to be regulated. We regulate software based on its impacts, based on what it achieves and what it does in the real world, and not everything of that is equally in need of regulation.

An application which runs on a single device, which doesn’t communicate with any other person and which doesn’t interact with sensitive personal information, essentially doesn’t need regulation any more than a notebook needs regulation – it has very similar properties. On the other hand, something like medical records management – where you’ve got security issues and personal data all over the place, where it’s networked, where the information may be transferred internationally – something along these kind of lines is extremely in need of regulation. So we have a range of degrees of requirement for regulation.

Bitcoin is of course the blockchain that everybody has heard of, and Bitcoin is, frankly, insane, it’s one of the maddest things to ever happen, and I mean this in the sort of very literal sense. You have a piece of software that issues currencies and makes payments, already this is strange; the currency is a currency that is contained inside of the software system, occasionally referred to colloquially as “magic Internet money”, strange again; €25 billion has emerged from this magic Internet money, that’s a lot of magic Internet money; and nobody knows the original authors of the software. This is really something out of science fiction, that’s bizarre, this is completely unreasonable, but it’s the world that we live in, and we should get used to seeing unreasonable things.

Our model of how the world works is based on our own past, it’s based on previous rates of change; as the rate of technological change accelerates, we will see more and more and more things which are unprecedented and have this quality of “I don’t even have words for that, I can’t even actually name what that is.”

Faced with this kind of rate of change, we have basically three strategies: we can figure out some way of pulling the utopian potential out of these technologies, we can fail to pull the utopian technology out of these technologies and get a mixed bag which will tend towards dystopia, or we can basically knock the technologies back as far as we can and hope that we will get some kind of stability and stasis in the face of this rate of change. Obviously I prefer the utopian potential, but if it’s a choice between utopia or dystopia, maybe we’d prefer stagnation, and that choice is really at the heart of regulation of these kind of technologies. Can we avoid the dystopian future by anything other than stagnation? This is a heavy question; this doesn’t just affect blockchain, this is all of the new technologies. VR, AR, self-driving cars, drones, AI: that entire family of technologies are all going to push these kind of questions to the forefront over the next maybe five years.

If there’s one thing that I want people to remember today, it’s this: the future is a foreign country and it’s our largest trading partner. Normally, when we think about comparative advantage for trade, comparative advantage is between two countries that have different properties: one country is good at making wine, one country is good at making pork, you have an exchange and everything is fine. What we have now is comparative advantage between the present and the future. The present has an awful lot of real estate, the future has an awful lot of technology, and we’re continually having this trade between the present and the future. As William Gibson, the science fiction author, says, “The future is already present, it’s just not evenly distributed,” and we trade between the future forward and the future past parts of the world. Some places like London trade on their history, some places like California trade on their futurity, and the trade between the past, the present and the future is the majority of the actual trade. If you think about this in your own life, the building is 150-200 years old, the cell phone is six months old, and the day job is the thing that connects the two together – present, past and future in a trading equilibrium – and understanding that is a very important part of thinking about how we approach new technology. If you think of the future as being a coherent trading partner, when you see a new technology, the question is “Well, if this was being exported by another country, would we want it? How would we approach it?”

The other thing is that the future has no coherent representatives. There is no government of the future, there is no embassy for the future, so we have trade with this thing that breaks out right, left and centre, that brings completely new goods and services to the table but has no political representation, and in a sense we ourselves have to be the political representation of the future. We have to decide how the future will be perceived, how it will be represented, who it is that speaks for the future, and indeed who it is that speaks for the present and the past. I think we can all agree that the representation of the present and the past is generally much stronger than the representation for the future, but I think that’s changing as time passes.

The first thing that comprises blockchain is it’s a peer-to-peer network of computers. We’ve seen peerto-peer networks before, there’s nothing dramatic about a peer-to-peer network. The second thing is the peerto-peer network’s securely shared synchronised state. You’ve seen this before: this is Dropbox, this is Google Documents and any number of other systems, some proprietary and some open source. A set of rules for updating the shared state from a given moment in time, how do we know what a valid future condition is – again, no magic. Finally, the whole system is a software platform that’s fine, we have lots of software platforms.

In theory, there is no fundamental magic to the blockchain: it’s a set of components that we’ve seen before, arranged in a novel configuration. Skype, BitTorrent, Kazaa, LimeWire; we’ve all seen peer-to-peer, and society did not break down.

Google Docs, Dropbox: again, we know these technologies and we use them – no drama.

A set of rules for updating shared state is interesting. If there’s any magic in this system, this is where it is, because what we have is a much more sophisticated set of rules for maintaining the shared state. The rules for modifying the shared model of the world inside of this Bitcoin ledger system are much more sophisticated than are typical inside of these systems.

Finally, regulators are already used to managing software platforms, there’s nothing fundamentally new about managing software platforms. If all of this stuff is relatively familiar to us, if all the components are familiar, what is it about this particular arrangement of components that causes the drama? The answer is €25 billion appeared from nowhere. The only reason anybody is interested in this stuff is because it appears to be a machine that prints money. If the total market capital of all the blockchain technologies together was about say $100 million, I don’t think we would be here talking about this. At the end of the day, much as we all like to say it’s not about Bitcoin, if Bitcoin wasn’t going anywhere financially, it would not be about blockchain either.

Where is the money coming from? The answer is pretty simple: Bitcoin takes this technology platform that we call blockchain, and it uses that technology platform to implement what amounts to central bank of the Internet. This central bank of the Internet is a bit rough and ready, it doesn’t do all the things you’d expect a central bank to do, but it does kind of sort of work; it works, because the shared state represented by the Bitcoin ecosystem has a set of rules for updating the shared state that correspond to what a central bank would do.

We start at the beginning of time with zero Bitcoins, we have a monetary policy which issues Bitcoins at a rate of roughly 10 coins every 10 minutes, we have a set of rules for how those coins will be distributed, and we have a set of rules for transfer of the assets from one person to another person, and that very simple ruleset gives you monetary policy and a very crude, very simple banking system. That system prints coins, the coins float in a free market against other currencies, and currently it is issuing about $3 million a day. That’s where the money is coming from; it’s coming from exactly the same place that money comes from when a central bank prints money: nowhere, it appears from the ether. This is an extremely dramatic and peculiar development. You shouldn’t really have a central bank of the Internet; now that you have one, it’s pretty obviously a discontinuous change technologically, and now we have to decide exactly what’s going to happen to the central bank of the Internet. Maybe you just legitimise it, maybe you decide this stuff is not okay, but it has to be understood that this is the central bank of the future. The future has come calling, one of the things it has sent us is a currency, and now we have to decide how we’re going to approach that, almost as if it was being issued by a sovereign entity called “the future”. This shift in perspective, to understanding that the future is a coherent, other place from which new things emerge is the core takeaway today. Learn to think about the future as a place which sends us goods, and we can begin to understand the gravity with which we have to regard our relation with the future.

It’s not inherently surprising that you should be able to make software represent a central bank, because all of the existing central banks are made of software. This is not fundamentally dramatic. It’s a little strange in the way that it actually works, but the fact that it works in this way is just emblematic of the further direction of change. The evolution will tend to be in the direction of this extremely surprising discontinuous structure – things will come from nowhere.

The final closing of the loop is that Bitcoin uses the money that it’s issuing to pay for the services that are required to maintain the Bitcoin software platform. Essentially, it pays its workers in the cash that it issues, and then in this way it’s also somewhat sovereign, it’s somewhat state-like. It has a certain strangeness at the centre, where you issue the currency that you use to pay the people that do the work of issuing the currency.

Some very clever people did this, but although this is a completely new thing, we shouldn’t mistake the ruleset that Bitcoin instantiates, this central bank of the Internet, with the underlying blockchain technology. The blockchain could support a wide variety of rulesets; this just happens to be the one that they picked. This is a set of rules that a group of people chose to implement.

We don’t know who those people are, we don’t fully understand why they chose to issue that set of rules, but it’s an application that runs on the blockchain. The blockchain is the underlying software platform, Bitcoin is an application that runs on that underlying software platform, and it happens to be central bank of the Internet. This is not the only thing you can do with the underlying platform. If a platform is powerful enough that you can run a central bank of the Internet on it, you ought to be able to run large chunks of global trade, you ought to be able to manage the supply chain, you ought to be able to trade electricity. You could do just about anything with that platform, and the confusion between the platform technology which we now call blockchain, and the application which happened to be called Bitcoin, untangling that confusion has taken us about five years in the media, and even now there’s still a lot of misunderstanding of what’s connected to what. The thing I want you to understand here is that we can make the rules that run on the blockchains that we control. We can set up a set of rules, we can implement those set of rules, and it will run as faithfully as Bitcoin does more or less, without us having to buy into the Bitcoin political vision, which is a very peculiar one.

Now, this ability to implement your own sets of rules can be done in two basic ways, and the first is one set of rules per blockchain. This is expensive and inefficient, because you’re replicating the platform infrastructure every time you want to set up a set of rules; it would be like writing your own operating system every time you wanted to write a new application – it doesn’t make any sense. The project that I was involved within the blockchain space is called Ethereum. I have launched Ethereum, I was one of the initial communicators on the project, did some of the project management, so be aware that I have some skin in the game here. The critical thing about the Ethereum model is that the smart contract allows you to build new applications that share the same underlying blockchain. This has two advantages: the cost of rolling out the application is much lower, and secondly, the applications can communicate with each other to form systems, and those systems can be multijurisdictional. One team creates contract one, one team creates contract two, one team creates contract three, and those systems interact while they each have autonomous governance, and this is quite useful, this is an interesting thing.

Ethereum also has a central bank function much like Bitcoin, but whether or not that will turn out to be core to that technology when you look back in 5-10 years is hard to know. I generally am of the opinion that the coin systems are a temporary anomaly; not sure that we’re going to be mining these things in the future. Enterprise Ethereum Alliance is building versions of Ethereum which are not dependent on the coin systems, that do not require a central bank to function. They run much more like standard application platforms, they’re essentially cloud infrastructure, and blockchain on a cloud-like underlying system is a very reasonable way of proceeding. There’s no reason not to do that for the vast majority of the applications that people want to write. We do not need a blockchain that pays for itself using a central bank; maybe you do for some instances, but for the vast majority of things this is not necessary.

Let’s talk about future directions. The first thing is that the current blockchains which use this kind of mining model are slow, 30 transactions a second for Ethereum, and this is too slow to run the world – it’s barely fast enough to run a videogame. Obviously this has to change over time, there’s an enormous amount of work being put into making these things fast, but it’s not here yet. The second thing is errors in code are universal. If there are human beings involved, you will have mistakes, and in blockchains these mistakes tend to be irrevocable, so there’s a lot of work on using formal methods to mathematically verify the contracts to what you think they do. That work is at a relatively early stage, but there are a number of start-ups working in the area and some of them are making real progress. Scaled blockchain merged into the cloud. One possibility is that in a few years cloud and blockchain will become more or less synonymous. Cloud services will have picked up all kinds of cryptographic techniques from blockchain, blockchain will have picked up scaling techniques from cloud through technologies like lightning networks, and it will just become a general computing surface. Similarly, computer security on blockchains makes an awful lot of sense. You could easily imagine software update mechanisms merging into blockchain, so you simply have a blockchain-based mechanism for deploying for example antivirus software. All of these kind of diffusions of these technologies into each other point to a future where you have a single, well-defended, heterogeneous, parallel computing surface, a kind of distributed global supercomputer that manages a lot of the security processes and that potentially manages a lot of things we need, the supply chains, personal identification and similar. Exactly the shape of that platform and how long it will take to create nobody really knows, but this is the general direction of travel.

The blockchain hype cycle. The blockchain is essentially is as to Silicon Valley as Silicon Valley is to the rest of the world. Even people in Silicon Valley look at this and are just like, “What are these people doing?! This is crazy!” and it’s kind of nice to see Silicon Valley looking at some piece of new technology in total fear, it’s quite reassuring that they can still be surprised. Right now we have a full-scale hype cycle circus around blockchain, there’s no denying that. Large companies are trying to cash in, there are any number of scammy, weird little entrepreneurial shops running. Something very large is happening, but there are also all kinds of weird institutions attempting to grift on the energy of the blockchain, and differentiating what is good practice from bad practice is very difficult. Find deeply-technical people that you trust and have them guide you in this, because judging whether something is legitimate or nonsense based on its PR is basically impossible at this point. The hype cycle has gotten to the point where the PR materials from the scams and the PR materials from the real technologists are almost identical, so we are at the point where you need technical guidance to be able to distinguish what is viable and credible from what is nonsense. What is viable and credible is likely to wind up running the world; what is nonsense is doing a very good job of pretending it is.

Let’s talk about what the large-scale impact of this is. Right now civilisation is basically defined as an interlocking network of rulesets, governance systems. Having software that has the ability to automate rules and automate compliance with rules is inherently very interesting, because it potentially takes one of the core operating costs of society, which is creation and maintenance of rule structures and compliance with rule structures, and it offers the possibility of automating it. You could potentially partially automate away the costs of running civilisation, and this might be as big a transition as the transition from say barter to money. The idea that once one time we had to manually remember the law and that all of this stuff was stored on paper books to future civilisations might look crazy. “What, isn’t the law a piece of software where you describe what you want to do and it tells you whether it’s okay? – No-no, the law used to be on paper and we used to have experts in law. – Wow, that’s bizarre!” That sounds very strange and ambitious now, but the idea that you could digitise law in the same way that we digitised maps is actually not that unreasonable if you talk to the artificial intelligence people. The possibility for mechanising law and mechanising compliance should be taken very seriously as a potential huge breakthrough in our fundamental economics – it’s probably worth a few percent of GDP.

Similarly, you could think of all of these blockchain start-ups as being essentially rule entrepreneurs. “I come up with a set of rules for managing an electrical grid, here’s our solar, the solar will be represented by coins, the coins will be exchanged so that the amount of solar is never oversold, and we have a compliance framework built into that.” These are rule entrepreneurs, they are making up sets of rules, supporting those rules with software and then launching these rulesets into the world essentially as freelance governance. I think that that’s extremely interesting, the idea that we have an industry which is governance-as-a-service covers the vast majority of the blockchain start-ups which are out there, and I think government should be very excited about that, because potentially it gives government the ability to offload a lot of the work of managing new technology onto new technology. Now, making sure that citizen interests are properly represented in those processes is going to be a big, difficult job, but if we can just about regulate the banking system, we ought to be able to keep these start-ups in line.

One of the other issues is that these rule entrepreneurs tend to be global by default. Unless you specifically work very hard to stop people getting access to your ruleset if they live in another country, the rules have Internet scope. This is where we begin to see a lot of grinding pressure between the inherently globalist nature of anything deployed on the Internet and the national nature of the vast majority of the rules that we’re attempting to manage. That grinding surface between the kind of implicit Internet globalism and national interest is I think one of the most interesting areas that we’re going to be dealing with for the next few years. It should also be noted that most of the companies who are rolling out these kind of infrastructures is small. They’ve got the same basic shape as web design firms, not the same basic shape as banks, so you have small teams of people without deep legal expertise rolling out rulesets that are intended to change the world and thinking in terms of global platforms, and that mismatch between the ambition of these entities and their legal depth is causing a lot of the confusion in the blockchain space. But, how much entrepreneurialism do we want? If we want aggressive entrepreneurialism, we’re going to have to accept there’s a certain amount of chaos, and where that balance is is a political question. This is a thing for regulators and legislators to decide: how much chaos can we tolerate?

The thing that we’re attempting to prevent is enormous meltdowns. If we wind up with Ponzi schemes which wound up with half a percent or a percent of global reserves in them, you can imagine that instability in those things would become very dangerous. If Bitcoin’s value climbs from being say €25 billion today to 2-3 trillion tomorrow over the course of say another year or two, at what point do we decide that it’s just too dangerous to have that much money slotting around in a bucket with these attributes? Is there a point where you begin to apply deflationary pressure to Bitcoin to try and deflate that bubble, or do we accept that global trade is just going to be demarcated in funny money? These decisions have to be made, they have to be made by global consensus, and it’s very hard to know what the right decision is. If we’re facing global challenges on climate, maybe we need a global currency to face global challenges. If so, should it be issued by Bitcoin, should it be issued by the UN, should it be issued by the IMF, the World Bank, or should it not be issued at all? Huge questions of very, very deep political value and political philosophy that we’re going to be forced to make decisions on relatively rapidly.

Set a robot to catch a robot. The thing which would be most useful in all of this is if we had mechanical representations of law, it would allow people to figure out whether what they wanted to do was compliant or not compliant very, very cheaply. This is an enormously powerful new growth area, and the more work that could be done on creating software definitions of law that would allow people to automatically check whether their plans are compliant, the better the future will be. I think the idea that we have to dramatically reduce cost and the incidental costs of complying with law to make it easier and easier and easier for people to actually stay within the circuit seems to be one of the most exciting things which is happening now. A lot of work is happening in Singapore, the term is smart statutes, but this is something that I would encourage over almost all the other things that we could be thinking about: how do we get the law to catch up with the technology so that we bring the cost of compliance down, so that we can have innovation at the same time as we remain compliant?

Globalisation 2.0. Pushing forward on this compliance attribute, this compliance angle, if you have well-managed repositories of digital law, it ought to be possible to cut the cost of compliance so that we essentially remove non-tariff barriers while retaining local control – global trade, local regulation, computers handle the red tape. The idea that you could represent what happens at a border in the software, so I can basically say, “If I take this product and try and move it down this route and sell it in this market, where will I run into trouble?” and the software tells me, “Your label is wrong, and you’re going to need to pay this tariff,” then I could look at that process before I have to actually put any goods onto a truck. I could essentially simulate a behaviour, and then figure out whether that behaviour will be compliant with the law, whether it will get me the result that I want. If we had a world in which there was a lot of law represented in this way, it would dramatically facilitate trade, particularly when countries change their policies. Because you could change your policy, the computers would be updated, everybody would understand what has shifted as it affected their businesses, essentially automatic notification, and then businesses could adapt and life could continue. I’m hugely in favour of internationalisation by virtue of digitisation of law, and I think that we will look back and see that this was an inevitable trend. Progress is initially slow, but in the long run this is as important as port infrastructure. The idea that you have a national representation of your legal codes that will allow people to test the product or service against statute before they enter the market, this seems something that will be, with hindsight, very obvious.

This leads us towards a concept we call Internet of Agreements. As things currently stand, we only have proper, full digital representation of credit card payments, and that has given us a 20-year growth industry. The Internet of Agreements model, if we bring the rest of the payment infrastructure online so that we can have stocks and bonds and all the rest of these things in native digital representations, how much economic growth does this create? But, how do we then represent national borders so that we don’t simply have Internet native financial instruments crashing through national sovereignty? This is that interface that I think is best managed with digital law. We have Internet-native instruments which are then rendered down into compliant forms by essentially legal oracles. This is very much a 10-year horizon vision, but if we don’t start taking proactive steps to bring the full weight of technology behind rule of law and behind lowering the cost of building an increasingly complex global architectures, we will discover that the technology is dragging us rather than us leading the technology.

I went to Dubai about two years ago and designed their national blockchain strategy. The national blockchain strategy of Dubai was very straightforward: issue all government documents on a blockchain so that the government can register and understand its own documents. Dubai then put a $200 million fund behind that vision and behind their general vision of proactively engaging between government and the future, all run through an organisation called the Dubai Future Foundation. This sort of an approach is then proactively building a trading relationship with the future. They want to become a kind of future port where ideas are imported from the domain of potential into the domain of reality and then exported from Dubai to the rest of the world. That kind of future port infrastructure is quite capital intensive, but it also has enormous benefits in terms of Keynesian spending and improvement in education and skilled labour availability. So even though most of the money spent importing ideas from the future is wasted, it is only wasted once, because on the second pass you get a skilled labour force. It’s very important that we think in these kind of terms, because if we don’t get better at pulling things from the future to the present, other countries will continue to accelerate and we will lose our position.

One of the possibilities is that these are the first waves of technology to which a genuine panEuropean response is possible. As one of my co-workers, Rob Knight, says, “If we were implementing mobile phones today, Europe would simply have its own country code; you would dial 90 or whatever it would be for Europe, and any phone in Europe would ring.” We’ve got that opportunity with blockchain; there is no need for every single sovereign entity inside of Europe to build its own blockchain technology. There’s an opportunity for shared technical infrastructure at a pan-European level which would be an enormous binder together of the European bloc, and I don’t think there’s any reason to have the inefficiency of myriad incompatible national systems. It would really be the first technological change where Europe has been fully stood up and fully ready to actually build at that kind of level in an almost supranational way.

The deepening of the single market: again, relatively obvious as a principle. If you have the ability to do things like track and trace for physical goods at point of manufacture, that system could be pan-European, and it could give the European trading bloc a set of technical capabilities that would be unique in the rest of the world. The ability to fully understand what your goods are, to understand where they came from, to understand where they’re going is exactly the kind of subject which would serve underlying European political agendas like environmental concerns, but without putting us in a position of having to do sort of brute force legislation. We build the platforms at the pan-European level, and then European business simply builds out onto those structures.