EB135 – Stephen Palley: Lawmodynamics And How To Sue A DAO

Transcript

[Music] This episode of Epicenter Bitcoin is brought to you by Hi.Me. Protect yourself against hackers and safeguard your identity online with a first class VPN. Go to hi.mme/epicenter and sign up for a free account today. And by Ledger, makers of the best hardware security devices. Have peace of mind in knowing your private keys are protected by industry standard physical security. Go to ledgerwallet.com to see the full range of products and use the code epicenter at checkout to get 10% off your first order. Hello and welcome to epicenter

bitcoin a show which talks about the technology projects and startups driving decentralization and the global blockchain revolution. My name is Brian Fabian Crane and I'm Mroy. Today we are going to talk to Stephen Paley who is an uh who's an attorney based out of Washington DC. Lately he has been writing a lot about Dows uh the legal aspects of Dows and and other topics. I found all of these blogs very interesting. So I'm I'm really pleased to welcome him on the show. Before we start, Stephen uh perhaps a bit about your background. Sure. Uh thanks very much fellas. I'm glad to be here. Um I'm a lawyer in private practice in Washington DC. I've been practicing for um about 18 years or so uh both in the

Midwest and on the East Coast. My practice is historically focused on insurance coverage for policy holders and particularly in the construction industry. And I've always had an interest in software and software development. Um, a couple of years ago I had a crazy idea for a piece of um um I guess you'd call it a negotiation uh platform uh dispute resolution software and I worked with another lawyer to build something um and in the process I didn't really know much about modern development tools or modern coding practices learn enough about modern uh development tools particularly Ruby on Rails um and about uh software as a service in general eneral to I would say

uh maybe become a little bit dangerous or I suppose framing framing it more positively. Um to appreciate the wonderful things that software developers and engineers do and to be able to have more of an intelligent conversation and in the process my practice has moved that way in part to working with people to develop what I call um compliance driven software. And as you guys have pointed out, I also apparently can't keep my mouth shut. And I write a lot. And um I'm glad that a couple of people have found some of the things that I've written um to at least be notable or interesting enough to object to and um glad to be part of an interesting conversation.

Yeah. So for for listeners who have never read his blogs um you might want to go to uh so all of all of the blogs are on LinkedIn I think which probably is not one of the platform of choice in this community but uh you can you can go to LinkedIn and check out all of the articles that talk about how DAO and like like you know the the the legal uncertainty surrounding it and then there's another article which lays out one of the principles Stephen has developed which is called the first law of law dynamics and we are going to walk through that that concept as well. But Stephen, how did you get interested in this field? What what brings you to our part of the universe? You mean uh blockchain, smart contracts in general?

Um well, I I also because I'm a lawyer, I have to give a little disclaimer. Um I although I'm a lawyer, none of this is it's kind of irritating like if you know me, you can almost say it for me. None of this is legal advice and I'm speaking. These are my own off-the- cuff opinions. Um, none of this is approved or agreed to by any law firm that I might be associated with. And if I have any clients who are watching this and they disagree with what I'm saying, I'm sorry, but I'm not speaking for them. Um so how how I got interested in blockchain smart contract it's it's a um it comes in part out of an interest of uh making law more efficient and automating things and a couple of years

ago I ran across a fellow named Casey Coleman um who a lot of you know who's now with ARIS which they've had wonderful success um for which I'm I'm very very happy and I think I probably heard the term smart contract contract first from Casey and if he were on this he would point out very quickly they're neither smart nor are they contracts but I became fascinated with the idea of process automation which is part of what the project I worked on which is it's defunct though but still lives on in code impassbreaker was about the issue was how do I take settlement negotiations which are ultimately about finding a bottom line that everybody knows to and instead of spending months and months or years and years nickel and dimming one another is there a way to

automate that Um, and I think that whether you're using a distributed ledger or not, whether it's blockchain or not, what is incredibly interesting to me about the technology and about code itself, is how it takes um how it provides the ability to automate performance. And I think that's that's what has gotten me interested in blockchain and smart contracts. And um I think the Ethereum uh world or community uh is um particularly interesting because there's so many smart people who are working in it and who are interested in this taking processes complex processes and putting them into code.

And um I don't think it ends up getting rid of lawyers necessarily, but I have to tell you, I mean, if that happens eventually, I think I may have said this last night, if anybody's out there has read Herman Hessa's the glass bead game, Magister Ludy, like if it if it all means that eventually I'm coded out of existence, there are no disputes and we get to sit around in white robes playing math games with glass beads, I'm okay with that. So, I don't see it as a threat. I actually see it as an augmentation. I'm not sure that always comes across. I think a lot of lawyers probably feel that way too. Was that that was a very long answer to a short question which I think I'd probably be guilty of frequently this evening. Cool.

Well, we we certainly love that. So, you mentioned performance here. One of the phrases that is often used when describing a lot of these things that we're doing here and that blockchains and smart contracts accomplish is the idea that they remove kind of the requirement to trust or they uh they are trustless. Um what do you think of this term and does that relate to what you mean when you talk about performance? Yeah, that's interesting. Um, did you mention the first law of lammo dynamics a second ago?

We just might have. Yes. Should I should I talk about that? Please do. So, or am I jumping too far ahead? So, um, I'm really interested in the idea of trust because it's something that is part of negotiations. Um, and I'll give you an example as a little bit of a war story which lawyers are guilty of. This is involves the first case that I settled as a baby lawyer years ago. Um, and the other guy, the guy on the other side, let's call him uh, his name was Wall-E, and it was a $10,000 property damage claim. I've been practicing law for maybe two months. And I had, I don't know, two $2,500 on the table and he was at 12,500. And I said,

well, why don't we just like we both know I was like I knew I was like, the case settles for $7,500. But we were like going 100 bucks a time. I was like, "Well, why don't we just why don't we just go right to $7,500?" And he what he said to me was I I'm pretty sure he said, "Kid," he said, "Kid, you got to dance the dance." And the reason you got to dance the dance is because nobody trusts each other, right? Because if I put my cards on the table and it's 7500, how do I know for sure you're going to, right? So negotiation is all about there are ways to get around that, but it's all about trust.

So the idea of a trustless system is one where it's sort of what I initially set out to design is where you can remove that you can create a system where um I don't have to trust the other side because something neutral that doesn't have a bias and doesn't care is going to take care of it for me. Now the first law of law dynamics says however that um risk or trust they don't you cannot engineer them out of any system they go someplace I think um you should stop me if you want at any point but I was going to go to the story that we were talking about before we went on the air so when I hear trustless the words that the word that

comes to my mind is riskless And I think about um something that happened to me about what year is it? I don't even remember what are we 2016. Yeah. So this would have been about I don't know 10 years ago. Um, and I before that, several years earlier, I'd spent a lot of time in a very quiet anonymous office in the middle of the United States in a very low-level firm handling some very insignificant, except of the parties involved, real estate litigation, and I went through lots of loan files. Um, and these cases were very sad and they often involved people losing their houses. And I looked through these files and I saw

things that I didn't understand. I saw um parties called nominees. I saw something called MS, which I think people know of as a disease. Other people know of it as a was basically a clearing house in the United States that was set up to repackage loans, make the transfer of mortgages easier. I'm oversimplifying greatly. Anybody who knows what it is is probably cringing now, so forgive me. it uh but it didn't really make any sense but I figured I was I didn't understand something and what I came to understand was part of this mechanism had been come to exist in order to or in order to support a um mortgage securization business uh which had taken on a life of its own

and I had a conversation about 10 years ago with a fellow in New York who was with a really big fancy investment bank and he was talking about these things collateralized debt oblig collateral collateralized debt obligations mortgage back securities and he told me that they were um riskless and I like was explaining you take a thousand mortgages you put them into this into an instrument um you create a trust you have insurance you overcolateralize the holders of the debt can never lose money it's like and I asked him I said what happens if a bunch of the mortgage holders can't pay and you need to foreclose And he looked at me like I was an idiot who crawled out from goodness knows

where and didn't know the first thing in the world about finance and I should go back to the provinces blah blah blah. And he said it's riskless. We overcolateralize. We used financial engineering. We created trunches. We've got insurance and two years later everything went kaput. So when I when I hear trustless I think the same thing like the risk didn't go it didn't disappear. It went someplace else in the system. And it's the same thing with trustless um uh any sort of trustless trustless peer-to-peer uh currency smart contracts. It might be that we can take the need for you and I to trust each other out of a system, but

we put the trust into math into something else into minors. So that's what I think of when I think of trustless. I would also ask you guys, I would put it back to you. Is there any advantage in calling something like Bitcoin or Ethereum trustless? Do you need to just just before before that I I'd like to ask you one more question on that. So I I I get your point that the trust doesn't go away and I agree with that. I think that's true. But would you agree that the trust can the requirement for trust can be reduced so and risk can be reduced?

Um well reduced from what? Um, can we locate it? Can we move it? Can we identify it? Can we centralize it in an algorithm? Yeah. But think about it like there's an interesting analogy with electricity. Um, and I I can't remember where I heard this. It was an interview some years ago. Basically, what this person said was, look, I mean, you can distribute the risk from power generation by having plants all over the country, right? coal, gas, whatever. And that creates a certain environmental footprint. Uh, but if something goes wrong in one plant, it might be bad, but it's not going to be catastrophic. Or

you can focus all of that risk in a couple of nuclear power plants where you have a you're able to control it greatly, right? You have a lot of control over the risk and you can use engineering to manage it, but if something really bad h something bad happens, it's always going to be really bad. So I don't know if you can reduce it. Maybe you can. I think you can locate it though and maybe you can remove it from you and I and you can put it into an algorithm or you can put it into miners or you can move it to cryptographers or cryptography. But I'm not sure that it is reduced. I just think it moves. Having studied chemical engineering uh actually uh we we kind of have have a version of this uh I'll try

to explain like the first law of law dynamics in using an analogy from say building a dam or building a dyke. So, so imagine you have a river that you know that is an unstable river. It changes course every year, right? The Yangze in China is like that, right? One one year it's flowing one particular way, the next year it will flow somewhere else. It keeps shifting. Now the the issue with rivers like that is if you build your house next to the river, next year it shifts, your house is going to get flooded. So if you build your house in the plains of such a river um you have risk that someday the river is going to shift and it's going to flood your place. Now so in like imagine like very

old times you know we face that risk humans face that risk. Now then came the invention of dikes and dams. Uh what what with dikes and dams what you can do is you can take a river like that and you can force it to always go into one path. Mhm. You can you can you can build structures that you know uh always force it to go one way and so it it doesn't keep shifting every year. Right. And so the obvious the the thing that humans perceive is that the risk has reduced the risk of flooding has reduced.

Now what happens when people perceive that the risk of flooding has reduced? People build a lot of houses in that plane right there. They're going to build like earlier say there's about a thousand houses now there are going to be like 100,000 because everyone perceives the risk is reduced. But what also happens concurrently with this reduced risk is in one once in 100 years there's going to be so much rainfall that there's going to be such a huge flood that the river is going to break the dam or the dyke. Right? The human uh structure is not going to be able to take all that water and it's going to be broken and then the river will flood the complete plane and it's

going to kill everyone who built a house around it. So the you reduce the occurrence of the disaster but at the same time you actually increase the magnitude of the effect when the disaster happens. So in a sense like the first law of law earnings I think tries to capture this right like we can reduce the the risk of some bad thing happening but then all it does is maybe maybe the disaster is going to come later and it's going to be way bigger and if you if you multiply these two the risk of a disaster multiplied with the impact of a disaster then that is kind of constant.

All you can do is uh all you can all you can do is uh you can go low risk but high impact but you can go low impact but higher risk but that multiplication factor kind of stays constant. Exactly. And that's a beautiful analogy and I guess what I would say there is when you recognize that like engineers don't design riskless structures. What they say is we're building a 200year dam, but you need contingency plans and you better maintain it. So, it's the same thing what I worry about in talking about something or people can call things trustless if they want, but I would very respectfully

suggest that it's important to think about contingency plans and also to think about engineering principles. I mean, how do you know according to whom? What are the risk factors? what happens if something goes really bad. Um there was a recent piece um published by R3 uh by uh by Vitalic Buteran. One of the footnotes actually talks about um a potential risk associated with trusting miners. And look, I'm not a math expert or a crypto expert. I'd be foolish to try and um tease that argument out, but it's either I think it maybe footnote one or footnote two. It does intelligently which is not unexpected raise the question of uh where did the risk or where did the trust go in this

uh in this supposedly trustless system. I don't I don't think there's anything wrong with saying trustless for a particular purpose but not trustless completely. Um so that's that's where I get uh that that's where I go u when I talk about the first law of law dynamics. basically what I think what you've just described um maybe better than I did. [Music] Let's take a short break to talk about H HighMe. You know, you need a VPN provider to protect yourself against those nasty hackers trying to steal your private information. With HideMe, it couldn't be easier. You just install their application on all your devices, iOS or Android. Log in and you have a cushiony, cozy tunnel in which your data

can move freely and unencumbered, all the while protecting you from those needers. Now that's boom. To sign up for the free plan, go to hightme/epicenter. The best thing is when you use that URL, if you ever go premium later, you're going to get 35% off. And premium comes with unlimited bandwidth using up to five devices at the same time. You can use all their servers worldwide. You can pay with Bitcoin. And best of all, it comes with a feeling of peace and satisfaction like having tea with the Dalai Lama. We would like to thank hike.mme for their support of Epicenter of Bitcoin. I think there's kind of two sides to this. I think on the one hand, you're certainly right that the term

trustless is kind of misused with Bitcoin. This was always kind of super obvious in the sense for me like if you look at mining, right? Then you trust that those miners aren't going to go back and do a 51% attack, right? Which is something that's very hard actually even to assess. what's the probability of that because you don't know them. You don't know exactly know what their incentives are. So I think saying that the term trustless is misplaced makes total sense. Of course also you know who actually audits the software that is being run you know basically nobody or very few people except for a few projects. So there's there's risk there and there maybe risk on the hardware side and so

there's there's all kinds of risks but I don't think it makes sense to say just there's some sort of constant quantity of risk and it just gets shifted around. Uh I do absolutely think you can't reduce uh and change the the magnitude of that. So, and I guess it's an interesting question know this whole endeavor this blockchain project is that something that is trying to accomplish that right to create a reduction in the amount of trust required maybe that's one way to think about it but but I I don't buy the point that it's just sort of it's always there just

and it's always there in the same quantity and is just sort of morphs where it appears. Yeah. I mean, so look, I'm being kind of literary and sort of maybe too cute by half. I I I acknowledge that. I don't know how you mathematically quantify trust or risk. Um I'm I'm sort of trying to make a a point perhaps by exaggeration, which is whether or not you can quantify whether or not you can uh mathematically quantify trust. Um I'm not sure you can get rid of the need for it someplace. And maybe it's better to put trust in an algorithm though I'm not sure. Um it depends on the algorithm and it depends on the standards that we're using. Um I don't what I like about

um automation whether it's a crown job whether it's something written using Ruby or Solidity or it's blockchain or not is um predictability and process automation. Um and the same thing with that the thing I like about um distributed version control like git it was like when I saw GitHub for the first time it was like magic to me. Um you know the history uh you know the revision history. Now I know you can play with that right um but I love the idea of revising things that way. That has nothing to do with I'm not sure that I need to call that a trusted or trustless system. To me those aren't

necessarily helpful words. Okay. So, we've spoken a bit about um you know trust and that that doesn't disappear. Now, kind of maybe the the flip side of this is the question of liability, right? So, things things are going to go wrong. Smart contracts are going to lose people's money and a lot of these people are now developing contracts that then they will run themselves. So how do you think about the question of liability in these systems?

So the world's a really big place. How many countries are there? 180. I don't know. So let's say there are 180 countries. Maybe I'm wrong. Um and there let's say in the United States there are 50 states. I'm only licensed to practice law in four of them and I only have active licenses in three. And the law is different in all of them. So I'm necessarily going to speak I'm going to speak in a very generic sense. Part of the answer is you know it really depends. it. I used this analogy last night at the Ethereum meetup. Um I think people may find that response irritating until if you're a developer, if I were to ask you, what's the best language to use? What's the language that's going to um be less uh like it's going to allow me to create something that fails less

frequently? Is it going to be C++? Is it going to be Haskell? Is it going to be Ruby? Is it going to be Solidity? What should I use? Should I use it? You're going to say it depends. And sort of the same thing here from the standpoint of coming up with single answers to questions. So having given given you that really long caveat which is probably irritating uh let me see if I can give you some general guidance. The way I look at it um so code in and of itself is just that it's not sensient. It doesn't have legal existence. It's not a person. um it could become a person if it's a part of a corporation which in most places is recognized as sort of a as a as an artificial person.

If a programmer creates something that is flawed or bulky in the United States um they could and um their flaw um is um an actual and proximate cause of harm and they cause someone damages. they might be liable in tort under a negligence theory. It's possible they could be liable under a strict liability theory under products liability law. Um it's also equally possible that there could be a contract claim if they were um in privity of contract with someone.

Um it's um open source may be different. You may not have a contract, but under certain circumstances, you may be deemed to have a duty or an obligation to not do things that cause a lot of harm. Um, so here's an example. I sent mayor a copy of this article earlier today. It's an incredible article. It was apparently funded, this research was funded in part by um, oh, what's his name? Elon Musk. And it's called um, what was the title of it? How to create a malevolent AI. I mean, it sounds crazy, right? I mean, it's a wonderful article, scary and hilarious at the same time. And you know, basically what these guys say is, so there's been a lot of talk in the

security community lately about preventing things, but like we don't really know what would happen if there was a malevolent AI, so maybe we ought to think about creating one so we can know what to do if one shows up. Um, be kind of like creating a supervirus to test. I'm not I'm not passing judgment, although it sounds like I am. I don't want to be forcibly cyborgized, which was one of the parts of their risk matrix. That would be bad. So, I'm good with preventing this stuff. But, I guess, you know, it's a roundabout way of saying what would happen like what would the theory be if somebody created an open-source malevolent AI, pushed it up to GitHub, and the thing caused a lot of harm. I think there are in a common law system, I can imagine lots of ways

to go after those people who probably only meant to do good um but maybe ended up doing a lot of harm. Part of something to think about too is just because you do something for free doesn't mean someone can't sue you. Um you know sometimes it's the stuff you do for free that causes the greatest harm. So, kind of going back to a sort of more narrow response, um, if I were a developer and I was creating things and I was giving them away, I probably would think about what could go wrong because things can go wrong. And I would not assume that I'm going to be scot-free or or free from liability. I think there are

ways to uh to be liable. Um, does that a specific enough answer? I feel like um I don't want to be too vague. Uh, so feel free to beat me up a little bit and make me be more uh more clear. Oh, it it it makes complete sense, right? Like like let's let's let's take a few examples, right? So the simplest examples we might take is um the Ethereum Foundation, right? Th those guys went and created GU and uh and that GU software basically kind of underlies the whole Ethereum network at least today, right? And nobody in their right minds, none of the authors in their right minds would ever say

that this software is always going to work in all situations, right? they there they there could be consensus bugs. There could be things that lead to a fork in the chain that leads to like nasty scenarios and people end up losing their money or losing their sleep or whatever, right? I mean nobody would in their right minds would deny that this possibility exists. So uh I I I I I actually like agree that you know there's always the risk and as as an engineer you always kind of accept that whatever you have created has that risk right now the the hard hard problem is when when somebody's making something open source and you don't know who is going to end up using it because suppose

I'm I'm creating Ethereum or the next cryptocurrency I don't know then I I don't know who will end up using it and they could be spread across all of the jurisdictions of the world, right? How do I and in principle I accept I accept my uh I accept the risk in what I'm creating. How do I kind of ensure myself against liability? Yeah. So there's a concept in US law and I I'm not an English lawyer, but I think the the Brits follow it too. Maybe they don't.

And I I shouldn't speak of British law, but um at least in the tor context of reasonable foreseeability and duty, at least in the tor context, if you're giving stuff away, there's no contract. I guess the question would be um to if you created a malevolent AI and gave it the ability to attack life, health, and safety systems and actually wrote that code. I don't know, maybe it's reasonable to and you didn't bother building any security so that um nobody you didn't bother building an off switch. I can see a TOR claim against you. Let's say you wrote a um I don't know some sort of better CSS, right? Like you managed to like you you

all of the beauty of uh or maybe not beauty, but all of the utility of um um what do you call it? Um Bootstrap, but in 10 lines, 15 lines, I don't know. You're a genius. And somehow there's something about the code that causes somebody's um application to crash someday and because it's been incorporated into I don't know what somebody bonks their nose on a door. I mean that's different right? So you have to look at what the application is uh maybe what's what the reasonably foreseeable uses and um what what could go wrong. That's kind of the way I would look at it. like

what's the range of what could go wrong? Um how could people misuse it? Um and I'm not saying that giving stuff away to people who you don't know necessarily is going to mean you're liable. We just we don't know. So you have to I mean part of it is um use common sense. I mean it's one of the it's a lawyer's best friend, too. Uh, if you build stuff that's really na nasty and inherently dangerous and you give it away and it's used for a bad purpose and someone dies, I mean, maybe you should be maybe maybe you do bear some responsibility or liability for that. I'm not sure there's anything wrong with saying that. I think a lot of I think a lot of very um smart and

responsible programmers would uh would agree with me. So there's no silver bullet to this, right? Like there's no there's no method left. There's no process by which like I I build something and and you can and some somebody can come to me and say okay me do steps 1 2 3 and four and this will guarantee that you will never be held liable for what you have built and give it away. You know, if I could predict the future and I could tell you the law of every country in the world, I might be able to do that for you. Um, I think the best that you can do is um, it's a matter of probabilities. Um,

and there are ways to I mean, again, it comes back to the the it's impossible to create something that is riskfree. Do you ever read Nasam Talib's black swan? The black swan. It's a great book. And basically what he says in that is essentially in boiling it down is the only the only certainty is uncertainty. And it's a question of managing that. Um and I think that's true for um every profession. Um if you don't mind, I know we want to go on to uh talk about other things, but he uses an example which I think is wonderful. He goes to uh Las Vegas and he talks to a casino manager and um they talk about risk. And the casino manager says basically, I'm paraphrasing, he says, you know, we

don't actually worry about um making money or losing money. We understand that, right? We we get how this works. What he said, what I worry about is the things I don't know and I can predict. Like for example, the person who fails to fill out a license application to the state and causes us to almost get shut down. So, I there's no risk manager out there in the world, I think, who would tell you that you can come up with a perfect hedge. If there was, I mean, I think I'd probably become I'd be a very very wealthy lawyer, right? Cuz I would have the answer to everything. Um, we'll talk afterwards, right? Today's magic word is collision.

That's C O L L I S I O N. Head over to let's talk bitcoin.com to sign in into the magic word and claim your part of the listener reward. [Music] We've been speaking kind of about smart contracts in a in a very general way, but now there is a whole bunch of activity around the concept of a DAO uh decentralized autonomous organization which does have some interesting features, right? Because you can really kind of neatly separate the creation of the code and the operation of the project of the enterprise, right? So somebody can write the code and then somebody entirely different group sort

of goes implement it and then those people all may be totally anonymous they may uh they may not actually be sort of involved in the implementation itself as it kind of runs on Ethereum of course DDIO which we've done now a few episodes on already uh being the most uh famous example of that. What do you think of that? is I I know you've also written about some of the the legal risks around that. Yeah. So I don't want to single out any particular DAO big or small. Um what I would say is that I see one of the words I I mean I knew the word before but one of the words I learned from developers early on in my more

recent journey in this world is iteration. Um and software development as an iterative process. Um I think that's very useful. So, I don't think it's fair to criticize a the a DAO framework uh whether it's called standard or not for not being perfect. It's version one, a version two. You have to look at it for what it is. And um I see the thing about the DAO that I think is interesting and useful. Um, I think you see some of this in Kristoff Jensen's paper, which is um, the notion that you can automate governance. And I don't certainly don't speak for Kristoff or

any of those folks, and they might not share my view on this, but the as a lawyer, the interesting piece for me is how that code can automate what an organization does and how it operates. and by doing so remove the need for trust between people because the the relationships or the relationships inter say between themselves is placed in code. I I think it's a very heavy lift to expect at the outset to be able to take an entire governance structure and put it into code. It's complicated. Um, and the um, in our presentation last

night, I think it was Shawn Murphy, a very fine lawyer at um, and excuse me if I've gotten his name wrong, at Norton Rose, um, described the DAO as um, I mean, I I think of it as you've got DAO code, which is dry, sit someplace, then you have the DAO once instantiated, it's instantiated by someone, then you have the DAO um, as the relationship between people, which is I think what the way Sean describes desried it. I think it's very very useful. And so what I I know there's been um there's been some heated discussion and conversation um online among people about the best form of

governance. Um I've been impressed by, you know, aside from some of the things that have unfairly been said on Reddit about people, I've generally been impressed by the level of discourse even when people aren't agreeing with each other. I think what you have to do is continue to iterate. Um that's um that's something that lawyers do and I think that's something the developers I think it's something that we have in common. Um, but so to focus a little bit more specifically, you wrote once in one of your articles that when you hear some of the language of the the DO DAO, what you're thinking about is well, they're avoiding form, you know, forming a sort of a traditional legal structure. But

the implication of that is just that the court is sort of going to assign you one. Uh and the the one that you mentioned there was the the structure of a general partnership where you said then people would be kind of liable. [Music] Um first of all you know that that would have very serious implications. No. And is there a way around that? So very simply um it's consistent with what I've always said. um if you do not create rules or an organization, it is possible that in any venture, a court will impose those rules um upon you. Um I'm not saying that any

particular DAO, any particular place in the world has to have any particular organization. Um I would I'd be providing legal advice to people if I did that and I'm I'm not going to do that. Um, I'm also not going to speculate about um what any particular court is going to conclude about any existing DA. Here's what I'd say though, something that I said last night. Um, this goes back to the idea of the risk goes someplace. Um, if I were getting involved in a venture that ultimately had people involved as stakeholders, I might like to know what rules apply. And some people may be comfortable thinking that it's enough to have some smart

contracts controlling things. Um, another way which might not be right for everyone to manage the risk associated with being a default unincorporated association or partnership. That might not apply in every country in the world. But one way to deal with that is create a standard corporation. um have um a meeting um have a resolution attached to which is some smart contract code which and that smart contract code would um include certain governance rules. Have an old-fashioned vote of an old-fashioned entity to place a portion of your governance on a or the

blockchain. Everyone votes, everyone agrees. Boom. have an old-fashioned entity which has um newfashioned newfangled governance. Now the response that I've gotten from some people and it's fair I'm just this is that um I'm missing the point the beauty of automation and that people don't need these entities anymore. Um that's not for me to say. Um I'm just saying the risk is if you don't have an entity, something goes bad and your DAO decides to create a malevolent AI and decides to rob a bank. Um, you know, it's possible that you could have some risk or some exposure. One way of managing that is to pick a corporate form. That's not a perfect hedge. Incidentally, you can

break through a corporation under certain circumstances and get at the uh the participants or a partnership. Um, I think what's more important is to ask the questions unemotionally, rationally, and make a decision. Um, and maybe, you know, there's nothing wrong with asking the question and deciding that you don't care. I mean, you just have to be willing to face the consequence. That's up to it's not up to me. It's up to others. So, what I'd like to clarify for some of our listeners is um, so whenever like a group of people are are doing an activity and and they don't assign a corporate form to the activity, right? Then it's like an unincorporated association of

people doing something. Then then the then the then the challenge is that if somebody sues sues the group as a whole then the the law enforcement system can come after you individually. Right. Yeah. It's I think it's important not to generalize internationally. Um that can be a risk and that's a risk of a uh in the US um sort of a default general partnership. There's a that piece I wrote, how to sue a DAO, which I probably should have called how to be involved in a DAO and protect yourself or how not to be the guy in footnote 3. If you read the footnotes, I give a little war story in it about being in court and seeing this poor guy end up getting stuck with a lease that somebody else had signed that

he knew nothing about. And it was because the court found that they had a general partnership. Uh court looked at their conduct, their behavior, and decided that it was a sort of a default general partnership. as a consequence that party B was responsible for stuff that party A did even though party B didn't know about it. Again, I want to stress I'm not saying that is always going to apply to every DAO or that it applies to any DAO right now. Alls I'm saying is might and you might want to think about it. Um and there are solutions or at least you know know your risk. Um that's um associations are slightly different than partnerships under US law. And again in some countries, you

know, there might be an Italian lawyer or Estonian lawyer or a or a um Kenyan lawyer listening saying, you know, Pal's all washed up. He's got it all wrong. There are solutions under Kenyan law or Likensteinian law. Um and um you know, that's the the hard part. It is really hard. I I said this um this someplace else. Distributed presence can create distributed liability. And if you're a big company, let's say you're a Ford or you're Apple and you want to distribute your product in 180 countries, you probably are going to follow the law of each of those countries and get opinions in each of those countries, right? Um

it's expensive. It's kind of the way the world works. So if you don't do that, you take on a certain amount of risk. Let's take a quick break so I can take it to Paris. I stopped into the Ledger offices and met with Eric Lavec, Ledger CEO, and he filled me in on the upcoming Ledger Blue. In early 2016, we are going to release the Ledger Blue. This is a personal privacy device which runs on a secure element, has a touch screen, NFC, Bluetooth, and USB connectivity. And it will be a full-fledged hardware wallet with a seven factor validation of the transaction directly on the screen. It will be fully open source. You will be

able to add your own apps and it will also be compatible with Fido second factor authentication. Password has going are going to disappear and it will be replaced by this kind of cryptographically secure authentification. The Ledger Blue will be certified by Pho and will give you the possibility to login on any website very easily just by signing a cryptographically secure challenge. Ledger is making hardware wallets easy and convenient without compromising on security. If you want to get a secure setup for storing your bitcoins, go to ledgerwallet.com and use the code epicenter to get 10% off your

order. We'd like to thank Ledger for their support of Epicenter Bitcoin. So another topic that you know is somewhat related to the DA or can be at least but has been sort of in this space for a long time is the idea of of token sales and crowd sales and of course those could be either conducted by sort of projects that aren't really DAO like let's say for example the Ethereum Foundation back then or now they have been conducted for example by you know the DAO and I'm sure many many more will follow Um what is do you think those are legal? Are they you know specifically on the US law do you think they are

sometimes legal and what kind of makes a difference here of whether they are okay or not? Um it depends on the jurisdiction. Um I'm sure people who are familiar to the space are familiar with the Howie test. There has been plenty of written on the subject. Um I would not want to speculate about um securities law risk for any particular DAO. Um I would simply say that um those laws exist for a reason and just because something is digital doesn't mean that it is outside of the law. Anyone who thinks that is

free to think that. But it's uh it's what I call VHT uh or vacuous happy talk or delusional. Um so people are free to reach their own conclusions. I I I think it is there's a wonderful paper by I can't remember who wrote it. There's been some really good work in this space and I think you can do things that involve I mean a token I a token is a um um it's a database record, right? Sort of. All right. Yeah. Look, I mean, it's certainly possible to am to if it's a database record, database record isn't inherently anything. Um, the question is if you're selling people the database

record, what do they get for it? Um, and talk to a securities lawyer. That's what I would say. And, um, you know, be careful. Um, just because there's some I don't have the the site in front of me, but something I wrote recently. Basically, a judge said, it's like the judge basically said, "Yeah, I know it's new. We've dealt with new things before. I mean, the law is what the law is." And this is something that was true 100 years ago, 500 years ago. Um, you know, and, uh, the law did change in response to the printing press. You had the development of copyright law and the statute of an really interesting, totally off topic. Um, but it's not like there aren't laws on

the books. Maybe they'll apply in some cases, maybe they won't. So, so far there's been there's been a lot of those and so far there hasn't been any real action from SEC or others. Do you know does that mean they are okay with it or does is there just such a long lag time? I'm checking my telepathy. Wait, wait a second. like trying to hack it. Yeah. I I mean uh so I will say this generically and this isn't necessarily related to DAO or blockchain.

Um law takes a while. Um but limitations periods are long. Just because something doesn't happen in a day doesn't mean it isn't going to happen eventually. And again this is just a general principle. I'm not I'm I am not singling out any blockchain project. This is just a it's a good thing to know. Just because you do something and you don't hear response from law enforcement the next day doesn't mean that you might not hear something a year later. It depends what the statute is. There are some long statutes of limitations and you have people who have lots of things on their desk and they've got time and sometimes um you don't hear anything because people are investigating. I honestly

have no idea. Um I you know that's that's basically what I would say. There have been prosecutions involving Bitcoin of course and in the US um the decisions um which you may or may not be aware of the courts have said yeah Bitcoin is essentially it's a cash equivalent or it's currency right um so it's not like courts haven't been able to understand what Bitcoin is um and I think the same thing I haven't checked in about a month but the last time I checked US case law I didn't find any references to Ethereum um it's probably still the case. Um, but I mean database, it's a question like

what do you do with it? If you used Ethereum, if you used Ether to buy something illegal, um, it wouldn't matter that you'd used a distributed database entry to do it. It would still be, you know, it's likely that would still be a problematic transaction. I'm sorry I can't give you an absolute um answer. I just I I think anybody who tells you that they know for sure um is either lying or they they work for the SEC or you know regulatory agencies. Um anywhere I I um I do think like if it were me I would can I would look at the people who are involved in the project.

I would look at their track record. Um I would look at um governance. I mean you can tell a lot about things by the people who are involved and in by their um by their behavior. So So uh another topic we would like to go into is the is this notion of jurisdiction. Uhhuh. Right. So uh first of all like let's let's let's kind of have a rough idea of what is the notion of jurisdiction for courts. And the reason we want to go into this topic is we want to drill down into how a court could assume jurisdiction over a DAO's activities or not h how how this could

happen. So first of all but we should go into what what is the notion of jurisdiction for courts. Sure. And I'm speaking as a US lawyer about general principles of US law and it's been somebody there's some been wonderful comments online. U Americans always think that the rest of the world is just like them. I really don't I'm actually sitting in England right now, great country. Um I think the rest of the world is wonderful. I just don't know law from somewhere else. So I apologize. I'm not assuming that US law is co-extensive with the rest of the world's law. I mean it would be useful.

I could like give you blanket answers if it were true. So um you know again to my lawyer brethren elsewhere in the world if it works differently where you are um that's cool. you know, please say something. Generally speaking, jurisdiction is a power of a court to decide things. Um, and I wrote something about this recently and distinguish between different kinds of jurisdiction. You've got um ultimately you have to have subject matter jurisdiction, which means um the um ability either granted by constitution or statute to decide certain kinds of disputes. So, for example, the US is a federal system. And we've got federal courts and state courts. Federal courts are courts of

limited jurisdiction and their jurisdiction is provided the jurisdiction is provided uh by the constitution article 3. They're also statutes that grant jurisdiction. One juris one type of jurisdiction is called federal diversity jurisdiction. And in order to file a a case in federal court in the United States under diversity jurisdiction, um I have to have um at least uh pardon me, should have turned off Twitter. Mary, you're tweeting. Um you're multitasking. Uh in order to have federal court jurisdiction, you have to have um no plaintiff can be uh res citizen or resident of this of the same state as any defendant. So I have to be

like for example um if Steven Pali sues me Royo um if I live in Maryland and you live in Maryland I can't the court won't have subject matter jurisdiction over a dispute between us under diversity jurisdiction. Another component of diversity jurisdiction is the amount and controversy. Let's say you and I are citizens and residents of a different state but the amount in controversy is only $25,000 exclusive of cost and interest. if I sue you, um, there's still no diversity jurisdiction, which means the court doesn't have subject matter jurisdiction unless some statute applies. So, that's the first issue.

Does the court have subject matter jurisdiction? And there are lots of federal statutes that provide courts with subject matter jurisdiction. Um, another type of jurisdiction is personal jurisdiction or if you like Latin, impersonum jurisdiction. And in personum jurisdiction or personal jurisdiction is the right of a court or the power of a court which has subject matter jurisdiction to impose their authority on a person. Um if you're present in a state or in a federal district um you a court may have personal jurisdiction over you by dent of your presence and you can then be served with the summons

and complaint that way. If you do certain things in a jurisdiction, a court may by statute have personal jurisdiction over you and that's by something called in some states a longarmm jurisdiction. So if you go to Kansas and you commit a tort in Kansas or if you sell insurance in Kansas, not licensed in Kansas, I just happen to like it. It's not as flat as people think either. um you might be subject to a lawsuit in Kansas even though you're not a citizen or resident of that state and you're not present there but by doing certain things there you've in effect consented to personal jurisdiction a court can also have jurisdiction over things that's called inrem jurisdiction and that can happen if there is a piece of property that is

subject to a dispute but no person can be found and there are cases actually there are some cases uh which you'll see where Bitcoin has been sued. So those are sort of the first principles. First a court has to have subject matter jurisdiction. It's got to be something about um dispute that it has the power to decide. Then it has to have personal jurisdiction. And you can't create subject matter jurisdiction. I had a case involving a dispute between somebody who was I don't know I think they were in Bise and somebody who was in another state in the United States and they filed a lawsuit in federal court in Missouri and the court you know the lawsuit had a forum selection clause of Missouri and the court was like well doesn't have anything to do with

Missouri and I've got better things to do with my time so you can't fabricate subject matter jurisdiction so you know there are limitations um So like like this this this notion of jurisdiction seems to be like one of those themes that will come up uh once like I think I like for me I I think there are going to be cases people trying to sue Dows and and all and I think this this this notion of jurisdiction is is just very interesting because I think like u I think if if if there's some guy who's involved with a DAO and somebody else outside the DAO is trying to sue the DAO and get to this guy Then perhaps the uh the lawyer of the guy who is inside the DAO part of

the DAO can use this notion of jurisdiction to try to argue right that uh the court doesn't have jurisdiction over u over the defendant. Yes. Can can this this like this notion of jurisdiction uh seems to me like one of those things that's going to be fought out in court and clarified uh when the court have jurisdiction over it. Any lawyers who are listening would probably tell you this is uh you covered this in civil procedure in the first year of law school. I'm not actually saying anything that's very wise or brilliant. I'm just providing basic um it's what's known as hornbook law. These are really really simple principles. A lawyer doesn't have to know anything

about code to understand them. Now what's interesting about them is not to get too far a field. It's actually not I mean I'm kind of talking these things are um subject matter jurisdiction. You can think of it as a class and within it there are different subclasses that inherit. I mean it's very there things about this that are very very similar to code and there actually things about this that would be interesting to um put into smart contracts eventually because some of what courts do is decide whether they have jurisdiction. Um, and I think some of it could actually be automated.

Interestingly, I I don't think we'll see federal courts doing that in the next couple of years, but it's I think there are a lot of judges who would be perfectly happy if that portion of their uh judication was automated or simplified. It would certainly do a lot for docket control. other uh other things at the intersection of u say traditional contracts and smart contracts like structures that kind of fuse the two and create something that is useful are there like parts of that space like that are interesting to you so yeah so again I'm at the risk of um embarrassing him I probably wouldn't

listen to this anyway but I got this term from Casey Coleman and that the notion is dual integration where you have dry code and wet code and where you you you put as much of the automation or lattice work or framework as you can into software and um automate certain things. But there's certain parts of contracts that will remain outside of the software and that might be those might be you know general conditions, terms and conditions. It might be you might have a variable. Let's say you've got a a a strct that is um contract and within that you have a variable which is um um a contract and it's actually it's

or general conditions and it's a PDF. You're pointing to a PDF that is something but you've got you have variables like um you know uh starting date, ending date, daily payment, contingency. Those are all things that uh can easily and probably are already included in solidity smart contract codes. So that's a structure that I think is useful. And you know if I were you know and if I and as I think about building these things what I would do is look at traditional contract forms and start peeling away the things that can be performed automatically. But I would also think about ways of being comfortable with keeping uh certain

items in the um I think you would call it the wet code outside of blockchain. Everything can't be automated, right? I mean it's um it's um particularly things that are offchain. I mean they're hard. There's nothing wrong with that. Doesn't mean it can't happen eventually, but iterate, right? Take your time. Okay. So um so yeah like like this this theme is kind of uh one of the one of the things that is I think special to AIS industries like it's it's it's like one of the few companies that are kind of can in their blogs and all kind of come out with with with this view with this notion that uh you can have integration between like wet code and dry code and combined they can do

something that is that is much better than wet code or dry code alone and and I find this like a really really powerful notion because like there are many things in which you almost seem to need wet code, right? Like think about a mortgage, right? Um so uh I I I I go to a bank and I and I ask for a loan to buy my house. Now in this case uh a smart code just a smart contract is pretty useless because in the future uh I need to make payments into the smart contract because I took the loan but the smart contract does can't kind of if I if I stop paying it can't go out into into the real world and foreclose my home. So we do need like a wet vet code contract

for this interaction. But what where a smart contract might be useful is kind of automating all of these real estate payments. So like combining the these two things seems to be like a very powerful could be a very powerful area. Yeah, absolutely. And I know that's what um Casey and uh Nina Kilbride and Preston I mean they're all lawyers and um I know that's something they're very very focused on. So I'm excited to see what they come up with. Okay. So uh so one one one final topic we'd like to touch on is the is the topic of insurance and Dow's insuring something.

So you have been you've been dealing with insurance for the construction industry and insurance like is one of those industries that is very well very highly regulated right and um so just just tell us broadly why insurance is so regulated and what are the issues when kind of Dows try to ensure something? Yeah. So again, speaking of the United States, I I had a tour of Lloyds today. Um and that was a Thank you, Stuart Mast. Um that was uh quite something that Lloyds has been in business for several hundred years. I mean, it's 400 years, 500 years. Amazing to see. The reason that um in the United States um it is particularly difficult is because insurance is for the most part regulated

on the state level unless you're talking about certain types of group health and life um insurance. But for personal lines, uh you've got 50 state insurance commissioners each of whom have um regulatory authority. They may have approval authority over forms, over uh rates and premiums. Um why is it set up that way? There's a consumer protection uh function or purpose. The way insurance works generally speaking is um you know I can self-insure, which means I can put some money into a bank account if something bad happens to my car, then I'll have the money to pay for it. or I can put money into a pool which a third

party manages um and they manage it and they hopefully invest it uh prudently and providently. If there's ever a problem, I'll be able to then make a claim and get money back to cover my loss. Now, if someone doesn't manage the funds properly, um then I won't have the funds available to pay my claim. So part of what insurance regulators do is they um they help enforce, set or establish um reserving requirements, the amount of money that an insurance company has to set aside to pay claims. They also make sure that forms are fair. And I'll give you a simple example. Um and I'm

speaking mostly in the consumer context, but it can apply in other contexts too. So um one of the earliest uh form standardizations was through something called the New York Standard Fire Policy which is a statutory fire insurance policy form which um is used to ensure homes and it includes in many states a um sort of a guaranteed baseline limitations period. In other words, um you are guaranteed a certain amount of time in which you can file or make a claim and if insurance company tries to shorten it, courts are required to ignore it. So regulators also in certain circumstances they try and make sure that the playing field is

level and fair and that forms um that are filed um make sense and have language that's fair and even handed. I think it's important to remember too that um insurance backs up or um covers risk of loss for you know trillions of dollars probably in real estate. I don't think I'm exaggerating in that number. Um and if you have an insurance market that um is not consistent in certain ways um you can create you can create risks to that um

into a huge market. So there's a I think there are lots of reasons uh why um insurance is heavily regulated and also from a the standpoint of creating something that purports to regulate insurance. You're going to have a lot of or purports to provide insurance. You're going to have some very interested state regulators who will do their job and um come asking questions. Um did that answer your question? I feel like I rambled towards the end there. I apologize. No, that does that that does answer the question like the the reason kind of this this came up is um in some of the stable currency schemes that we have seen like currencies that hold their value against um let's say they

are roughly pegged to the dollar or whatever right we have had we have had like the the cryptocurrency community has needed to invent the notion of um trustless insurance right smart contracts that insure people in case the peg is broken or something like that. And uh and like you have one of one of your articles that says like okay this is an insurance service that has a lot of regulations and is it even a good idea to call this thing insurance in the first place? Yeah. So I had an interesting conversation with um very nice man is it run? I don't know how to yeah Christensen. And I I don't want to call out any particular DAO, but I don't I'm

not sure that the intent there was to create anything that is necessarily insurance. Um I I don't completely um and I haven't looked at it recently enough to be able to comment on it, but because my point was in my post, be careful what you call it. So I mean, if it's not insurance, don't call it insurance because if you call it insurance, there are a whole bunch of regulations that apply. That's my thinking. Um, if you're going to call something cheese, there are cheese regulations. Don't get mad at me for pointing out that they're cheese regulations. If it's not cheese, don't call it cheese. It's the same thing with insurance. If it's not actually insurance, don't call it insurance. It's I understand why people call these

things smart contracts, but it honestly confused me a little bit at first. They're not really smart contracts. They're not contracts. I'm not I don't think anybody's going to stop using the term and if anybody's throwing stuff at me right now. I apologize that it's nothing personal. I just I I find it the terminology is a little bit confusing and that's was my point about that in that article. My question was um is it really insurance? I'm not sure that it is. If it is um yikes, right? If it really is cheese and you're selling it in Pennsylvania and a cheese regulator hasn't given you approval, you're um you know, yeah, it's

no different. All right. Well, thanks so much for coming on, uh Stephen. That was really interesting talk with you about all these topics. I think they're topics that are going to stay with us for for a long time uh and kind of accompany uh this industry. And I'm sure it's going to be very interesting also to see when when some of these concepts and ideas and projects are going to get tested in court and in real lawsuits and battles etc. So uh thanks so much for for discussion and for your work and uh we will look forward to kind of following where where this space goes and where your work goes along with it. Thanks very much. It was a lot of fun fellas.

Um honored to be asked to speak and enjoy the conversation. Absolutely. So Epson Bitcoin is part of the LTB network. You can find our show and lots of other shows online bitcoin.com. Uh you can also for subscribe to our podcast any with any app. You can watch the videos on youtube.com/epenterbbitcoin. And uh we look forward to being back next week. [Music] [Music]