 Promote my platform. A lot of the people here presenting promote something. Mine is a conceptual talk. So it's not tech talk where they go into source code, nor is it promotion of my own platform. It's a conceptual talk. So maybe before I start, quick question, who knows what governance in general? Oh, there's someone almost asleep over there says, you know, it's okay. There's two people. Okay, sounds good. So governance in one sentence is basically a way of governing a system or platform or something, which means analogy, a good analogy probably would be like a business model. Think of it as a business model or a big picture of how things connect, how things interact, how things exist in whatever ecosystem or platform. Okay, just a few logos. Some people show logos. I mean, the most relevant from these logos here is this one, Prism. It's already considered one of the top blockchain economics and governance companies in North America and Europe. We have people with PhDs from Harvard whose supervisors got Nobel prices about market design, contract theory, and other aspects of governance. So this is probably the best you can have in terms of governance. So I'm basically representing them in this region. Yeah, and, well, I've been a startup. I've been a coder. I've been a startup twice failed, once exited. I advise in AI and blockchain, basically, among other things. And I help some ICOs. So a bit of an agenda. So we'll go with basics. I think there are some people here from Ethereum. They can always chip in at comments because I'm going to touch on Ethereum. So current state, I'm going to do comparison of Ethereum and Bitcoin because both are very famous. Everyone knows them. I'm going to touch on some contenders and theories, and then I'm going to propose two potential sort of futures, intelligent futures, which as of right now, none of them shows. And this is based on the fact that I also consult and advise in AI. Okay, so, yeah, by the way, if and when, I want to make it more interactive. So that audience doesn't fall asleep, actually tries to participate. So if you have a question, don't wait till the end. It's a long question, wait till the end. But if it's something quick comment or something, yeah? Yeah, so if you have something quick to say, say, don't wait till the end, till it's really late and everyone is asleep. Okay, so I'm going to break down governance into three main things. Incentives, coordinates. So incentives is basically in blockchain, you guys call it bounty, et cetera. But let's just call it incentive, a bigger picture name. Coordination includes communication and decisions. This is consensus, consensus forming basically. So in simple terms, this is what governance is about, right? Okay, so this slide I'm going to spend quite some time on. So here's the original. So let's take Bitcoin and Ethereum as of right now. Both are working on proof of work. How many of you know what's proof of work? Oh my God, I'm dealing with a lot of technical people, very good. So if you know what's proof of work, this visual is going to be a lot more interesting for you. And here's the source, right? So he compares how proof of work works and how Bitcoin and Ethereum work with, guess what, any Americans here? I'm not an American, there's two Americans here as well. Three, this is comparing with your U.S. Senate. So we have few stakeholders, right? So governance talks a lot about stakeholders because it implies there's a platform or marketplace, which implies there's more than one side, which implies how they deal with each other, how they communicate with each other, what are their respective incentives, and how the decision making and consensus happens on the systems, right? So have a look. Very good comparison. Developers are the Senate. They develop stuff, right? This is the Senate part. Now minors, minors are the judiciary because developers can develop stuff, but minors, they might decide to approve or not approve, right? And then you have the community that has some voting power, and you have the House of Representatives. This is basically the infrastructure, et cetera. So do you think this is an optimal system of governance? I know we have a few Americans here, but do you really think your system is actually optimal? Yours is a representative democracy that made sense in 18th century at that time. Say again. Yes, correct. Yeah, it's a very good point. It's supposed to be, but now everyone says it's supposed to be. That's originally how you got incorporated, so to speak. That's right. But so what's wrong with this picture? Any idea if something is wrong? Or even if you don't necessarily want to talk about your Senate, what's wrong with Bitcoin system? I mean, there's a few things that I'll comment on, but what do you think could be improved on? Say again. Yeah, that's one. Prove of work is expensive. And also scaling is difficult. Okay, let me ask a question from a different point. Do Bitcoin developers have an incentive to develop or improve the system? Yes. Why? Not always, but so their incentive is to make the overall Bitcoin value to go up. Right? There is no direct incentive for them. This is an overall ecosystems value going up. Right? Hence the reason Bitcoin is where it is, hence the reason that the room even exists, because it was supposed to be or is supposed to be an improvement on Bitcoin. Right? Developers have very little specific to them incentive to improve. Miners. What's the incentive for miners in Bitcoin system? Who knows? Miners get, what do they get when they mine? They get rewards, right? They get block rewards, which is 12 and a half Bitcoins right now. What else do they get? Huh? They get some transaction fees. That's right. And they also need the whole thing to go up, right? The Bitcoin value, because that's also in their thing. But like the gentlemen here are mentioning, proof of work is based on the fact that it works on electricity consumption, right? And people buy hardware. Someone from previous speakers was showing the hardware like that. So do you think, and this is a leading question, do you think miners have an incentive to move to proof of stake, at least in Bitcoin? Huh? Why not? That's right. Exactly. They have a lot of rigs. They spend tens of thousands of dollars on the rigs. Why would they do anything else? They bought this. It works well for them. They go into mining pools and that's great. Why would they move to... Proof of stake is better. But do they have an incentive to move? They don't. Hence the reason you have all these forkings, hence the reason everything else is there, right? Hence the reason that the room is there, right? So again, why am I saying all of this? Is because if you think about it, every... This sort of a big system has different stakeholders. Each stakeholder has a specific incentive of doing or not doing something. How does Bitcoin improve? Bitcoin has Bitcoin-improven groups, some emails, but does community get to vote on things that Bitcoin can improve on? Yeah? Only buy? Correct. Correct. Is there any direct way for this community to vote on Bitcoin improvements? No, not really. Do miners get to vote? So to speak, no. Miners get to hire or to decide whether to get some of the developers to do their version of it or no, right? Again, why I'm dwelling on this is because for you to understand that incentives are very, very important in any system. Bitcoin is extremely important about really improving the system. Miners don't care about improving the system at all, at all. Users don't get to have a say anywhere. The best part is users are the biggest chunk of the community. They can't really push this or that or that, right? They cannot. So the reason, again, I'm dwelling on this is because you probably know some of these names. Anyone who knows Tezos and disclaimer, I don't support, invest or otherwise endorse them. I'm just approaching them from purely conceptual, objective perspective. Anyone knows Tezos? No one knows Tezos. Go Google Tezos. Tezos is basically, think like Bitcoin or Ethereum or whatever current system you take, except everyone in the system gets to vote. On changes they think would make sense. If those changes, like any of you can suggest a change, a lot of the people can vote on it. If it passes, it goes into their test net for a while. Then if that works out, then becomes integral part of Tezos moving forward. That's one, right? One system at least users actually get to say in how the community, in how the system works and potentially propose solution. Definity is the same thing, but rolling back. So if you vote on a change that overall a lot of people like, you roll back and it becomes effective as a retrospective time, not now, including. So from some time till now on to the future. There are a few other systems. I don't know if you know Futurekey again. Encourage you to check. Anyone knows Futurekey? Okay, Futurekey is basically a system without going into history of how this was created. It's a system whereby voters or people or community get to come up with values. Transparency, accountability, I don't know, trust. And then markets decide whether these values increase the overall well-being of that community. So there is basically value creation by people and then there is predictive predicting markets, which are basically deciding which ones are being maximized and whether the overall makes sense for those values. Encourage you to check. Liquid democracy, who knows what's a liquid democracy? Oh, great. What is a liquid democracy? Politics, fear. Okay, that's one way of putting it. It's a good way. So everyone knows direct democracy. Direct democracy means you vote, everyone votes on everything on all decisions, everyone. There is no such a system, current modern system, probably accepting some places where they still have tribes or whatever. Then there is a representative democracy. I'm not going to point to the U.S. But representative democracy, you elect the representative who represents you. Whether that representative knows anything about what you care or your interest, that's beyond the point. Representative democracy has a lot of problems. So liquid democracy is what? It's something that's not black and white like either. The idea is that, number one, you can either directly participate in whatever decisions you want or you can elect the representative according to their competence, their interests, their values, whatever, whatever. So it's like a hybrid. And you get to choose. On these decisions, I want to take my own vote. On the others, I want to delegate it to someone who knows better. This is called liquid democracy. And then there is quadratic voting. Anyone knows what's quadratic voting? So Vitalik Buterin came up with one. I mean, this is not his invention, but he has a system that he's thinking of also somehow introducing on top of Ethereum at one point. The idea here is that because one-on-one voting, one person, one vote has a big issue. And I don't know who knows that issue. What's the issue of one person, one vote? Pulls and a sheet voting for lunch. The minority always loses. Yeah, so correct. So one vote, one person, one vote always has a majority or supermajority problem, tyranny of majority, if you like. This idea is very, oh no, no. This idea is very simple. Instead of, so because in one person, one vote systems, what you do is you're voting either I like or dislike. Right? It's very black and white. It's zero, one. In quadratic voting, you're voting how much you really care about something. It's a great, it's from zero to one. You buy votes, you literally buy votes. Let's say you care about whatever issue it is, about Cambridge Analytica issue so much, then you pay with whatever the token or the thing is in your system. And then eventually that gets paid back to you with time. So the idea is basically, depending on how much, how strong you feel about a certain issue, you get to vote with your wallet, so to speak. Right? Exactly. So some of this are alternatives. I encourage you to know about them because a lot of this quadratic voting is already, I don't know if you know Exim Chain. Exim Chain, Exim Chain is a supply chain system based on Ethereum, I think. And I think they're already using this. So some of the other systems are being used as well, but they are not widely known just yet. Okay, so my suggestion is, again, keep in mind, I've been consulting and advising in AI since 2011. And I'm quite new to blockchain, like I guess a lot of you. I mean, maybe like seven, eight months into this. So my thinking was that, because I know a lot more about AI, my thinking was that in every big system, where you have multiple stakeholders, complex incentive systems and communication flows, and you can see from industrial age, and I know some of the speakers talked about internet history, but if you look even further down, from industrial age, we have been trying to automate and optimize things. Right? I don't know if you realize, but there is a lot of AI in your daily lives, and I'm not talking about Siri. I'm talking about everything from algorithmic stock exchange trades, all the way to how airplanes land, all the way to Google, whatever you do on Google. And I'm not even going to talk about self-driving cars, but there is a lot of AI that's there that does stuff that humans cannot necessarily do at scale and at speed and at reliability that it's being done right now. Now, problem, and to me, it's a problem, and it's not actually, it's like almost by design. It's a new thing, and we're still having a very human way of doing things, right? Bitcoin, how many of you have read the original Satoshi paper? Okay, well, a lot of you, right? So if you read it, if you read it carefully, you probably realize, and again, there were a number of other competing, sorry, papers, not at that time, but a few years before, a few years after, the guy didn't realize it's going to go so beat, right? So this was an experiment that got mainstream for a number of reasons that a lot of other previous speakers highlighted. Now, it wasn't made to be, it wasn't optimized, so to speak. Now my point here and what I'm trying to propose is that we optimize it. Humans have optimized it as much as they can, but then we all have our own vested interest and our greed and whatever, and hence we are here with Bitcoin. Ethereum has a lot of improvisation improvement to do as well, at least in my view. So my suggestion was not having something like SingularityNet or whatever, you know? People use AI on blockchain interchangeably, but have these three things that I highlighted. So incentives, decision-making, and coordination to be done using AI. Why? Because AI can change, depending on what's the overall goal or the overall vision for that platform, can change without any bias or any agenda or any greed or anything else that humans have in a way that everyone benefits. This is what's done in a lot more complex systems than blockchain. It's just you don't know about it because you're not interested and it's done so well you think humans do it because you think humans do so well. So my suggestion was, and I mean, I even put some generic terms, right? Reinforcement, learning, generative adversarial networks and stuff because you can optimize and change incentive systems of different stakeholders in a system like Bitcoin or Ethereum. You can have optimal market design because it's a market or marketplace. Any of the systems are marketplaces because there's a lot of different stakeholders. They transact with each other, interact with each other. It's a marketplace. You can optimize it, make it efficient, make it scalable, make it really work well. And I think a previous speaker was talking about likes, right? So one of the things I've been consulting is gamification. Gamification is rooted in human psyche. It's rooted on a fact and there is behavioral economics and what not to bake it. So if we can incentivize different stakeholders in a way that everyone overall wins, why don't we do that? We should, right? So my suggestion was that the current systems already start thinking in terms of how we can introduce some automation and some optimization and take human sort of human side of it outside. I mean, Buterin has a blog post. You can Google it. Buterin governance, you'll see. He has a blog post at the end of which he says, we think the best is that we involve all stakeholders. He says that that's the end of his long article. And that's right. Let's involve all the stakeholders, but he doesn't have a solution for it. So that's where you have some people, some experts who do this, get a Nobel Prize on this, come and tell you what to do, right? So again, like think of how we can really optimize and really make it scalable. And that's one, right? This is governance too, at least in how I see things. Now, governance three. Anyone knows CAPTCHA? CAPTCHA. Now, this side doesn't know CAPTCHA. You guys don't log in in any systems, like Gmail, Facebook. You know, it's that image under username and password. Anyone knows the story of CAPTCHA? Great, one person. How many of you are Canadians here? No Canadians? Only one American guy knows about the CAPTCHA story. Okay. Okay, CAPTCHA story, if I recall correctly, and there is a link you can, or you can Google. Started with a guy, Louis von Anh, who he was part of one, I think University of Washington, I think, or it doesn't matter which university, he was part of a project that was digitizing medieval texts. And medieval texts are not exactly well-known. They're not exactly well-written. And there is a problem of someone needs to eyeball. You know, you can have all the optical character recognition, but the digital text is not going to be recognized properly. So he needed people to eyeball. So he said, it's either I hire thousands of people to eyeball, copywriters, pay them millions of dollars over infinite amount of time, because we have so much text. Or why don't we have this? This is beginning of 2000s when he came up with them. Internet is still new. Email authentication, we're still fresh. You didn't have neural nets that can crack captures, which you do now. So his idea was if we put an image of what looks otherwise illegible, three, four letters to people who are authenticating at once, adding another layer of security to an otherwise breakable, brute force breakable user password authentication. And at the same time, we make sure with an algorithm that whatever that's input by the user is actually 99% certainty correct. You basically crowdsourced eyeballing for free. Now why I'm saying this? Now think of it. Proof of work, like gentlemen there was mentioning, you were mentioning. Proof of work is what? What is proof of work in one sentence? It's calculation of hash function. Does it do anything useful? No. It's completely wasting time, energy and everything else. And hence the reason we're talking about electricity, etc. Now imagine if instead of churning your CPU or your rig for nothing, wasting it, I always give an analogy. I say you go 10 floors up, come 10 floors down, show me you sweat it. That's called Bitcoin mining. That's literally it. People get Bitcoin not because they solve problems in the world, they help, but because they buy rigs and because they churn their CPU. It's completely waste. It's waste of everything. My point here, imagine to have CAPTCHA type of an approach and I can help formulate how that can happen for proof of work. Proof of work is nothing wrong with it or proof of stake. Nothing is wrong with those. It's just what you're spending your effort, that's what's wrong. When Satoshi wrote the paper or when the paper came out, do you know how he got the idea of proof of work? Who knows where he got the idea from? No one knows where he got the idea. Okay, one person. Anyone has heard of hash cash? Hash cash. Can you tell us about hash cash? What is it supposed to do? Okay, never mind. Hash cash is this thing, came out beginning of 2000. I think 2001. The idea was very simple. There was a lot of email. This is dot com boom, right? A lot of email was being used as spam. So email spam was on the rise. This guy comes up with an idea saying, how can we make, how can we desensitize, if you like, the spammers? If with each email that you sent, you get a cost to your CPU, then I cannot send mail and emails because my CPU is gonna die or it's gonna spend so much effort. This is hash cash. And if you read the first paper of Satoshi, like a lot of you did, it says, it says similar to hash cash. It's not similar. This is exactly that idea. His idea was, okay, every time I send an email, in email servers, you have a little function that generates a hash with identity of the sender and it's tiny little, I think it was like 10 bytes or whatever, on top of your email. And verification of that, just exactly in your Bitcoin, it's very simple, much faster. This was hash cash. It became quite well known and still, I think it's being used by a lot of systems, but it's not mainstream. Like, not everyone uses it. And again, so proof of work came from that. This is the inspiration. Now, you take that which at the time made a lot of sense. You put it in terms of churning CPU, but on this kind of a scale. That doesn't make a lot of sense because you're just churning CPU, you have rigs, and then you get Bitcoin and there's no incentive for you to change. So, yeah, you just become more and more, you have more Bitcoin, but no one else benefits from that, right? So, if you had a system whereby your CPU is actually using, is being used to solve a lot of problems in the world. Everything from health to education to poverty. If you could derive a system like that, where proof of work is actually making your CPU use, I mean, this is all, I mean, full session is about open and community and stuff. And there are a lot of big problems that need to be solved, but you don't have CPUs. Imagine crowdsourcing your CPUs and you can use proof of work or proof of stake or whatever, but actually solving real problems. Wouldn't that be great? That would be great, right? Anyone thinks that won't be great? Yeah? No? Some people are smiling. Yeah, here. Please tell us your argument. I think you guys with Folding Coin is something like that. With which one? Folding Protein is a good basis for that. You're talking about Fold it? Fold.it? Finding primary bomb or something like that. Yeah, so I think you're talking about Fold it where they found the AIDS enzymes three dimensions. Is it that one you're mentioning? Something like that. Yeah. I mean, that's one point. So, okay, what he's mentioning. Anyone knows Fold it? Great. Do you want to talk about Fold it? Computations for programming, proteins can fold in different ways. Yep. And it's a incredibly complex computational task, so it's a primary route source. Yeah, so how did they crowdsource it? They had a bunch of people who are not specialists in this. They gamified it, basically. They put it out there. They crowdsourced it. Whoever who solves it, they made it look like a game. And in, I think, two weeks if I recall correctly, it was solved. Whereas previously it took 20 years, a lot of experts, and they couldn't solve it. So they found a three-dimensional structure of that enzyme that was escaping the experts for a long time. So they gamified it. So this, yeah, it's that sort of approach, right? You can crowdsource or gamify something for free to you. Otherwise, for free. A lot of people do it because for them it's minimal effort and not much of other resource. Spank change. It's called what? Spank change. Spank change. In yourself, you spank. I mean, I'm aware of your book. No, really? Spanks herself an appropriate number of times live. And this is part of the program. Is it imaginary? Yeah. So I mean, yeah. So it's very normal, right? You all should go and now check the spank chain one. You can also check Jesus coin if you are a Christian or there is shit coin and others. No, I mean, but fair point, right? So imagine if you could have approval work designed in a way. And again, there are a lot of people who can help or who can design this in a way that actually solves problems. So you're renting your CPU to do something useful, actually, whether that's for you. That's for a community for the world, whatever, whatever. And I had this discussions from with number of people, including some people for material from cyber networks. Everyone agrees. And they're like, so why do we do this? There's an incentive issue there as well, right? You have to incentivize them to do this. So anyway, this is basically my last slide. So I wanted to leave you with that. And yeah, and here is some context in case anyone wants to follow up with discussion and whatnot. So think of it this way. Governance is AI on current systems. And governance three is even more fundamentally changing the way proof of work or proof of whatever works, whereby that proof of whatever actually does something useful. That's what I want you to take away from this. Thanks a lot. Yeah, you have a question. Is that governance is really the mechanism for achieving consensus, right? Consensus achievement is one part of governance. Governance is the overall how everything works with everything else. Consensus obviously plays a big role in that. Sure, sure. The other aspect, the thing that I think is actually the most important is determining what the scope is that we have any business making a decision about. Yeah, that's also part of what you're supposed to define as part of the governance. You are defining that. You're also going to define what's the overall vision for that, let's say Ethereum or Bitcoin. What is that you want to achieve? You define that. From that, you can already see what kind of incentives you should have for different stakeholders to work together collaboratively to reach that. So you have to define that. You have a very strong point on that. One of the challenges that the democracies have, the reason why they inevitably fail, is that their scope grows too big. That's right. And governance schemes, consensus schemes, do not scale. That's right. They absolutely don't. And so the more you can restrict what it is that they are governing, the more efficient they can be in scale in that respect. So you may have a problem that everyone agrees this is a set of rules that which we want to apply and we're going to govern these for this. And it might have one consensus set or the different problem might have another different consensus set. Yeah, so good point. So what you're saying, if I understand correctly, is how do you basically ensure that everyone at the same time somehow is winning? Because everyone has their specific consensus based on their own specific incentives, right? Somehow. So I mean, that's the goal of governance. It's not necessarily just looking at one. It's looking at the overall picture and trying to optimize that. And no one says it's going to be the most optimal since the beginning, but that's a fair shot of trying because, again, everything. So one of my teammates, her supervisor, got a Nobel Prize in 2012. I don't know if you read the book called Who Gets What and Why? All-Rose. All-Rose got a Nobel Prize in 2012 on market design. This is the book on the marketplace design. Anyone who wants to know how to design a marketplace that has a fair shot of winning, you should read that book. It's about kidney markets. The guy designed analog. It's not like he has a platform. He designed kidney markets, which are very, it's... I mean, if you read that book, you can see everyone has very specific incentives. A lot of them are completely contradictory to each other. No one has any steady, I mean, they have communication flaws, but it's also broken. And he designed what became a very optimal system, got a Nobel Prize for that. So again, what I'm saying is because, like, there are ways of designing markets. There's a lot of system theory thinking about this, right? So you can look at a big picture and design something that's optimal for everyone, not just for one. Everyone has their own respective incentives and relevant consensus on decision-making system. But how do you make everything work together is what governance is supposed to do. Otherwise, it's this, right? Because right now, we are at the very beginning stages, both for the real man Bitcoin and everyone has this very narrow picture of their own incentive, their own thing, right? That's why... And I mean, that's normal. It's still new, right? So what I'm suggesting is the next and the next step on top of that. I think as it grows, we get very real ethical changes. Correct. As to what are valid and invalid governance things. And you kind of alluded to it, but I think one of the critical properties for anything that's going to be successful and when the last is the concept, Nicholas Tellev likes to call it skin and the game. Yeah, correct. How do you find... The skin and the game. Yes, correct. You have to have the skin and the game, correct. And so like, I think if you don't, if you find a governance model that does not have that aspect, I think it's quickly going to come into the world. Yeah, correct. And they, but almost by definition, you have to have all of them have the skin and the game of the bigger picture. Otherwise, they are not going to work together. Yeah, but that's what we help for. That's what we help with. Thanks a lot. I'm here in case anyone has any other questions.