Dominant Assurance Contracts | Alex Tabarrok, George Mason University

5:12AM May 8, 2021

Speakers:

Keywords:

people

bonuses

refund

contribute

kickstarter

contract

problem

case

pay

blockchain

equilibrium

dominant

threshold

question

public

funding mechanism

assurance

public goods

project

proposition

What are we going to talk about today? And well, last week, I'm not going to do the full onboarding of where do we get to the whole book until here. And instead, I assume that most people know where we are for this session, just to give a lot of time for our fellow presenters. But basically, in the last meeting, we already kind of skipped ahead and talked about open society layer five, particularly about Okay, what can we happen? What can happen if we privatized the layer of cooperation? And today, we're going to talk mostly about, okay, well, what is so exciting about crypto commerce, and particularly about contracts, and not about any type of contracts? But what is exciting about the assurance contracts, in particular about dominant assurance contracts? Why do we need them? And I think, you know, if you read the chapter, and so far, until we hit that, it's really about like, a lot about one on one cooperation and why and the things that that can unlock and, and, and why that is really valuable. And dominance shows contract kind of taken to the next level. And I really am so thrilled to have several work here, because he co authored or authored, the paper dominant shows contracts in 1996 already, so and so quite quite a while before Kickstarter, and a few other a few other mechanisms that we know of now and appreciate where even existence. So really, really, I think, early stuff that is finally coming to fruition. And I think it must be quite rewarding for you, hopefully, but you tell us all about it. I'm not gonna take any more away from you. And after Alex presentation, we do a quick, a quick q&a with your questions. And then we'll actually have two other presenters here who are currently working on proposals in that space. So we're trying to already put into practice what Alex has already been not really preaching, but at least suggesting since 1996. So Alex, thank you so much for joining, really, really honored to have you here. And I'll share more info about you in the chat. And without further ado, please take it away.

Okay, great. Thank you very much, Alison, I assume we're all good on screen sharing. I'm actually going to talk about two papers. The first one is the the Aurora paper, the 1998 paper, which Allison mentioned on the private provision of public goods by these dominant assurance contracts. And then I'm going to talk a little bit about a more recent paper with some co authors implementing this in an experiment, a economics experiment. So you can think about this as the private provision of public goods. You can also think about this as kind of a better crowdfunding. So let's begin with public goods. So public goods, as you know, are a market challenge. Notice I use the term market challenge rather than market failure, which seems to prejudge the issue. But there's certainly a market challenge. Suppose that, for example, we want to avert a flood, right. So we know with a simple kind of Prisoner's Dilemma kind of game public good game, that if everybody contributed to averting the flood, that would be great people would get these high returns 900 900. But if an if everyone does not contribute, okay, then you get this terrible outcomes 00. And you would think, Oh, well, since it's so much better if everyone contributes, that everyone else will contribute. Of course, we know that that is not the case, because there are individual incentives not to contribute. And I want to kind of break this down a little bit more finely than is often done. There are actually two reasons why people might not contribute. The first is the free rider problem. So if we think about you is being agent I on the rows, then you're better off, if other people contribute, you are better off not contributing, if other people contribute, the flood is going to be averted. And by not contributing, you get a higher payout. You don't have to do anything. And you get the benefit of the public good. So you get a higher payout. So that's the free rider problem that if other people are going to contribute, then it's better for you not to contribute. But there's also a second problem, which I call the assurance problem. The assurance problem is you're worried that if you can feed mute, when other people do not, then your contribution is wasted. So if you contribute and other people do not contribute, then you're trying to put your finger in the dam, and it's not going to work and you're just going to be blown away by the flood. So your contribution would be wasted. So there are these kind of two problems the free rider problem, the insurance problem. Now, we can think about this as a game. And let's imagine now that the town's mayor is going to change the game a little bit. The town's mayor announces that those willing to build the dike will beat at Town Hall. Work will begin, if and only if enough people gather so that the diet can be built high enough to avert the flood. Okay, so that's what the mayor announces. Now notice, we've now gotten rid of the assurance problem. So everyone's going to meet at Town Hall. And if enough people show up, then you're going to if you have enough to avert the flood, then you don't have to worry that your contribution will be wasted, because you're only going to contribute if enough people show up, right? If enough people if if people don't show up, you don't contribute, your contribution is never wasted.

We still, however, have the problem of the free rider problem. Okay, we've gotten rid of the assurance problem. But we still have the free rider problem, right? Maybe if people are altruistic or perhaps do want to be seen as altruistic, we might want we might be able to solve that problem. But we still have that problem. Okay, now the mayor is going to change the game a little bit again. Now the mayor is going to announce that the dike will be built if and only if everyone agrees to contribute, if and only if everyone agrees to contribute. Well, now we've actually gotten rid of the free rider problem. Because if you don't contribute, then the DAG isn't built. So you do not get a bonus, you don't get a you don't get to enjoy the benefits of the dike. If you don't contribute, if you don't contribute, it's not going to be built. So the only way it's going to be built is if everyone contributes, and then they get the high payoff. You've also eliminated the assurance problem. So now we've eliminated the free rider problem, we've eliminated the assurance problem. Still, however, there's actually not a guarantee of success. Why not? Well, if you think that you will, if I think excuse me, if I think that you will not contribute, then it's rational for me not to contribute. And if I don't contribute, then it's rational for you to think that I will not contribute and in fact not to contribute. So you get a self fulfilling prophecy, right? So we can this do not contribute do not contribute can still be in equilibrium. And in fact, if we have kind of some epsilon costs of contributing, then this equilibrium is is quite plausible. Okay? crowdfunding, which we're all now pretty familiar with, I can't kiss Kickstarter, they use the assurance contract method. Okay, so the assurance contract is now quite quite familiar to most people. And the way it works is no one pays until the total contributions exceed a threshold high enough to ensure completion of the goal. And crowdfunding, it's grown rapidly. There's millions of these campaigns last year, raised over 35 billion worldwide. But there are still some problems. It's been quite successful, but there are problems with it. And the main problem is most campaigns fail. Kickstarter success rate is approximately 37%. Okay. And you might ask, why is this? Why is this? one possible answer is that you just have bad projects. That's possible. And, of course, there are bad projects on Kickstarter there are bound to be some. Another possible answer is that we are getting to many of the inefficient equilibria. Right. Remembering insurance in this game, there are multiple equilibria and they do not contribute do not contribute. equilibrium is one of them. There are lots of equilibria in which people don't contribute. So maybe when we see that Kickstarter success rate is only 37%. Maybe that's because we're in this do not contribute do not contribute equilibrium. Okay. Now we come to the dominant assurance contract. This was sort of invented before Kickstarter, but we can understand it pretty easily, more easily now because of Kickstarter. We can also think I call it the dominant assurance contract. You can also think of it as refund bonuses. And here's how it works. A crowdfunding entrepreneur offers potential contributors, potential contributors, a refund bonus, if they offer to contribute, but the contribution threshold is not reached. In that case, the potential contributor gets their contribution back, plus a bonus. Okay. So if you agree to contribute, but others do not, you get the bonus. Okay, if others contributed and you do not they get the bonuses. So if you don't read The threshold, which is required to produce the public good, you get a bonus.

Notice that at least in this simple version of the game, it's now actually a dominant strategy to contribute. So we've taken a public good game in which it was a dominant strategy not to contribute. And with refund bonuses, we've made it a dominant strategy to contribute. Why is this? Well, if you think others are not going to contribute, okay, so if you think others do not want to contribute first column, then you want to contribute in order to get the bonus on the others, if others are contributing, then you want to contribute in order to get the public good, in order to push you over the line to get the public good, right. So at least in this simple model, it's now a dominance, right? It's not always good, it's always going to be an equilibrium to contribute, it's not always going to be a dominant strategy. That's what you have these two players or you require everyone to contribute. The interesting thing about this is that in equilibrium, the refund bonuses are never paid, the refund bonuses are never paid. So by offering to pay the refund bonuses in the event that not enough people contribute, you may get an equilibrium for everyone to contribute. And so you never have to pay the bonus. So the bonus is kind of has this amazing free free property to it. For those of you who there's a kind of a famous paper in economics, that diamond into vivid paper, on deposit insurance, and this looks at Bank runs. And the upshot of the paper is that if you have deposit insurance, then you never get the bank run. So you by offering to pay this thing on an out of equilibrium path. The thing you're worried about never happens, and you never have to pay it. So it's a it's a great again, when it in theory you get you get a lot for nothing. This is a free lunch. Okay, we all know to be worried about free lunches, but this is kind of, in theory, a free lunch. Okay. This then raises a bunch of questions, right? How should the refund bonuses be paid? should be like a fixed amount, you get like $100 or $10? Or something like that? Should it be proportional to your contribution? Maybe you get 10% or 20%? Something like that? When should the refund bonuses be paid? We have some flexibility here, I'm going to show it may actually make sense to pay them only to the early contributors. We have some design freedom, because this doesn't happen in equilibrium. And then there's the basic question, does this actually work? Okay, it's great. In theory, tab rock doesn't work. Okay, it doesn't work. Right. Okay. So we run some lab experiments. And we basically set up a environment, which is similar to what you might see in crowds in a crowd phone in a crowdfunding environment, like Kickstarter. The people in the environment have an option to donate to different types of the crowdfunding campaigns. Okay. They can give some and they can then give some more later. Okay, they can upward revisions. You could do this continually. Okay. There's always multiple, there's always at least two Well, there's always two fundraising campaigns. In particular, the details here are not too terribly important. But we have 10 subjects per group. They have each each run of the experiment last two minutes, there's a hard close. We know their values. This is what's important about running a experiment, we're running an experiment running a lab experiment, is that we're going to pay them based upon values which we assign them, so we tell them this is your value of the public good. Okay, so this is all, you know, we had Vernon Smith a couple of weeks ago. And so this is part of Vernon's best insight, that you can induce values. And so we induce these values for the public good. And then we can compare it with theory. So we know this is an unusual with these experimental Labs is an unusual case where you can actually test the theory. If the threshold is reached, as with crowdfunding, each player gets their value. If the threshold is not reached, each player who agreed to contribute gets a refund bonus in the experiments with refund bonuses.

We're going to test a bunch of different ones. I'm not going to go through them all. The baseline, there's no refund bonuses. There's some where you get a fixed amount. There's some way you get at a fixed amount of given too early, there's some where you get a proportional This is like 10%. And there's early and not early, and so forth, I'll show you them in a little bit more detail. Here's the first kind of basic result I want to give you. In the baseline treatment, without refund bonuses, only about a third of the projects, all of which should, all of which it would be efficient to fund all of which would be valuable to fund only about a third are successfully funded. So that illustrates the large amount of equilibrium Miss coordination, that that 00 do not contribute do not contribute equilibrium appears to happen quite a lot. Okay, because all of these would make sense. If they were funded, perhaps, perhaps just fortunately, or randomly, it's it's actually very close to the Kickstarter number. I don't know whether how much weight we should put on that. But it's actually close to the clicks Kickstarter number. Here's the first basic result, any refund bonus, increases success by about 20 percentage points, or 50%. Okay, so whether you pay a fixed amount, whether you pay a proportional amount, doesn't matter so much all of them perform much better than baseline, not 100%. Okay, so in that sense, where, theoretically, we should be at 100%. So we're not at 100%. But we increase the number of successful campaigns by about 50%. The larger refund bonuses not terribly. Surprisingly, they work better. So that fixed amount of six works better than a fixed amount of three. The early 50 works better than the later 50, the 20% works better than the 10%, and so forth. We can also see that that equilibrium is talking about 00, that this is relevant, because in the baseline, nearly 20% of individuals don't contribute at all. Okay, you have contributions zero, this falls in half when the refund bonuses are used. Okay, so in the baseline, you get about 20%, they're contributing nothing, okay? And that falls in half, when you have any form of refund bonus. So it really gets people not just to contribute more, but it gets people to go from zero to one. Okay. Do you use Peter TEALS language? Okay, now, I want to say a little bit more about a variant here, which is focusing on getting people to contribute early. And Benjamin Franklin had this kind of saying when he was asked, you know, he was well known for raising a lot of money. And he says, How do you do this, he says, well, in the first place, I advise you to apply to all those whom you know, will give something next to those whom you are uncertain whether they will give anything or not, and show them the list of those who have given, right. So your first contributors are really valuable, because you get, you know, somebody famous to endorse your public good. And other people will then give they want to be they want to affiliate or they say, Ah, this is something which people are coordinating on. Okay. We want to do that as we want to do that as well. So there's something to these early contributors, which appears to be especially successful. And maybe the signals public spiritedness guide signals, which ones people value, which ones you want to fill up with. You also see this with Kickstarter campaigns, you can predict which ones are going to be successful very quickly. So the the six the successful ones, they're successful early. Okay, which is somewhat surprising.

Yeah, and so the question is, can we kind of use this to get double, double our money as it were to increase even further? The advantage of these refund bonuses? Can we use the refund bonuses not only to eliminate the 00 equilibrium, but can we also use the refund bonuses to sort of amplify the positive nature of contributing early? Okay, so we're going to compare for this question, we're going to compare particularly to refund bonuses, to refund bonus schemes, their proportional refund bonus of 20%. Okay, and that is paid anytime during the two minute contribution window. And we're going to compare that with the P e. 20, which is early 20. Money, where you get the same 20% paid on contributions made during the first minute. So the first half, so you only get a refund bonus, if you make a contribution during the first half, okay? And what happens? Well, the baseline 43.5%, or successful goes up to 60.5%. With the refund bonus of 20% goes up to 67%, which is higher, not actually statistically significantly higher, but it is higher with the early refund bonus. So the early refund bonus appears to give you something you can sort of see this here, it's kind of addressing hope this graph is not too confusing. Here we have the first minute of total contributions. And here's the second minute of total contributions. This is the baseline. And so if you want to be on the red line, okay, that's where the project is funded efficiently. So you want to be somewhere on the red line. Okay? That's the success. Okay, so being successful is being on the red line. If being early didn't matter very much, then the successes would be distributed equally. But what we're seeing here is that if you're not above 150, that is above half of what you need in the first minute, then you don't get to success. So there's very few, which come from behind. So this is where you know, you have less than half what you need in the first minute, you could catch up in the second minute, but there's very few up here. So you don't really catch up. So all the ones which are successful, they tend to be successful, because they have high first minute contributions. This is the baseline. Now you introduce the purple, which is the refund bonus paid any time. And now you can see you can catch up. So you have a lot more successes up here. Okay. The early 20. You don't need to catch up. And you get a lot more down here with the early 20. Of course, you get a lot more early, and you get a lot more successes. So you see you, you end up down here. Now, why do these refund bonuses work? Well, they what they do is they make a potential contributor, pivotal. What do I mean by that? a contributor is pivotal. Okay, when you have a single person who would find it profitable to push the total contributions to the threshold. So at the very beginning, it's in nobody's incentive to pay for the entire public good. Then some people contribute, you get closer to the threshold. And then at some point, you get close enough so that there's one, at least one individual who it is now in their personal interest to push you over the threshold and produce the public good. And what the P e 20. Does in particular, is it makes people pivotal earlier. And that I think makes being pivotal more salient. Okay. Here you can see the time to the first pivotal contributor, or the baseline. Okay, it's going up slowly, it's going up, it's going up, the first pivotal contributor

is coming out about the 102nd. So remember these things last two minutes. So you go up to 120. So the first pivotal contributor really doesn't happen until quite quite near the end of the game. And then it's kind of a rush, right? There's like sniping and people are rushing and you don't quite know what's going on. With the refund bonus. Okay, you get more people being pivotal. But again, it's right at the end. So it's not it's not salient, doesn't give you time to think about it. With the early refund bonus, okay, you get not too surprisingly, a bunch of people being pivotal quite early, before the 62nd mark, and then you get a second pivotal bump at the end. So I think by making people aware of their pivotal pneus earlier, there's a bonus to the early refund bonuses. As I pointed out a couple more slides and then I'll finish refund bonuses in theory, they're never paid. But in practice, you got to pay them sometimes. Okay. We don't see 100% success rates. Okay. Our best case is sort of a 67% success rate. So there are some campaign failures and then the entrepreneurs have to pay out their refund bonuses is In the know, 33% of cases where there's failures, so you might worry, is this going to be too expensive? It turns out that for an entrepreneur, it's better to offer the refund bonus. Because the number of successes, given some reasonable level of profit, when you're successful, more than pays for the refund bonuses when you're not successful. Okay, you don't need huge refund bonuses for them to be successful. And with modest markups, 10 or 20%, the refund bonuses are profitable. Okay, so that's kind of really good news for these things. Because even though they're not perfectly successful, as you, as theory says that they should be perfectly successful, but of course you don't, you don't always get that they're still profitable, they're profitable. And p 20. is especially profitable, because you get more successes, and you actually pay fewer refund bonuses, because you're only offering them to people in the first part of the campaign. Okay. Those are the basic lessons on refund bonuses are on the dominant insurance contract. Let me just say, a little bit we had Glen while, a couple of weeks ago, and while and the talyc. And so he hits a have this paper, which I'm many of you are familiar with our liberal radicalism. And it's a different method, right of overcoming the public goods problem. And I just thought it'd be worthwhile just making the distinction here, because I think both of these things are quite useful liberal radicalism or dominant insurance contracts. They both do, they do slightly different things. So what the refund bonus scheme does, is it works well, at overcoming the contribution problem, it gets people to contribute, it overcomes the free rider problem. So if you have a public good, and you know what it is like a lighthouse, like how big should the lighthouse be? Okay, you can figure that out from technology, it has to be a certain wattage and a certain height to do what it's supposed to do. Or if you're going to build a dam, you know, what the size of the dam should be, or a fireworks show, you know how many fireworks you're going to have. But the trouble is, not know that the trouble is getting people to contribute to these things. Okay, you know what it is you want, you just can't get people to contribute to it. That's what the refund bonuses, the dominant assurance contract is good to doing. If the problem, however, is that you don't know how much people value the public good. So you want to you're trying to figure out how much you spend on public parks versus fireworks. That's more of an information revelation problem.

And that's where the quadratic funding mechanism booter in encik, and while that is very good at usefully revealing information, the negative is that it requires some government or outside funding. Okay? So dominant assurance contracts is good, where you know, what the public good is, you know, what you want, and you just want to incentivize people to contribute towards it. The quadratic funding mechanism is good, where you don't know what you want, you want to figure out what people really value. And you have an outside funding source, so you're not as worried about paying for the thing, because you have some outside funding. So I think both are useful. And more generally, what I think is exciting, right? Is we typically think about public goods as being outside the market mechanism. Right? So, you know, it's often public goods versus the market or government to produce public goods versus the market. Okay, as if these things were opposed. But refund bonuses, dominant assurance contracts, quadratic funding mechanisms, things like this, what they're suggesting is that it might be possible to have the market produce public goods. And we might actually get a lot more and better public goods, if that could happen. So these mechanisms, rather than being opposed to markets, markets might be the solution to actually discover more and better public goods, more of what we actually want. That's kind of exciting. Okay, so conclude the refund bonuses, they double crowdfunding campaign success. So we're trying to get Kickstarter or another crowdfunding campaign site. Let us do a field experiment. We have the lab experiment, we got the theory, we got the lab experiments field experiment is next. That would be could be very useful to a Kickstarter like platform. They pay for themselves. All kinds of refund bonuses could work fixed, proportional early concept they all work well. Early refund bonuses are especially powerful as the increased campaign success and they cost less. And finally, market provision of public goods may lead to more and better public goods. Thank you. Thank you very much.

And ethic. Thank you so much, Alex. Okay, I'll stop sharing your screen in case you don't mind. Maybe we'll hop into the into the slides. Again, I'm realizing I have to update much in the book. Now, after visiting, for example, you know, I, for example, said that even w insurance contracts require an external funding mechanism, just because someone needs to put up the refund, which, you know, I had thought is similar to the matching and quadratic funding a little bit. But you know, I think that the because or, yeah, I think that because they suggested the paper, maybe one could use insurance premiums paid as a funding mechanism there. So I think that that would be, I think, an interesting way to explore how those those events could get generated in the first place. But um, okay, we have Kate with an upholder question here first.

So I have two questions. Well, first, thank you so much for that presentation. That

was fantastic.

My first question is, can you explain further how a dominant insurance contract solves the free rider problem? In the the mayor's game, he had everyone participate? But there's no such requirement in the dominant assurance contract? That I understand. And then my second question is, why do you think people why do you think the experimental subjects went for the do not contribute? equilibrium, the 00 equilibrium? Were there any opportunity costs? Were there any transaction costs? It seems like all the reasons why in a real world scenario that might be in equilibrium don't really match the, the experimental scenario.

Right. Okay, so to answer your your first question. So, suppose we have a sort of proof by contradiction. Okay. So suppose we have a situation where we have not reached the contribution threshold, okay? Well, now, it's in someone's incentive to add a contribution. Because if they just add a small contribution, then they're going to get the refund bonus. Okay. Now, you think, keep going with that. So you keep going with that until we reach a situation where if somebody contributes, now, the public good is provided, they don't get the refund bonus, but they get the public good. So the only equilibria are ones in which the public good is provided. So anytime we have somebody give any time to public, anytime you have not reached the contribution threshold, then either you gain by contributing in order to get the refund bonus, or you gain by contributing in order to push you over the threshold. So those are the only equilibria which remain are the ones in which the public good is is provided. Now why, so this is true in theory. Second question is, you know, why doesn't it happen every single time? So I don't think it's an opportunity cost. These guys are in the lab, they have to be there anyway, they've sort of agreed, you know, the, I guess they could be playing with their phone or something like that. But they also have the they're making real money, right? They have a chance to make real money, the man and they make more money, when either they get a refund bonus, or, you know, the public good is provided they make more money by playing then by not playing. I'm not sure why, like, Why doesn't it work? I mean, maybe it takes people time to understand the situation. There's still some, like, there are still tactically multiple successful equilibria. And there's some gaming perhaps about which one of those successful equilibria you end up at, like all the equilibria involve the public good being provided. But there's still some coordination to find out which one of those equilibrium I'm just not Yeah, I'm not totally sure. Like, you know, the theory. economic theory often does not work perfectly. People it's it's the people's fault. They're not rational enough.

All right, thanks. What was super interesting I think in the paper also that like smaller prizes, and smaller refund prizes worked better than the larger ones until I think like that paper was 70 credit.

And so the larger ones would work better than the than the smaller ones, but but you don't need huge ones you don't need really,

yeah, sorry. That's what I yeah, you don't need Like insane, insane prizes. Mark, you had a question too?

Yeah. So far, we seem to be only examining the dangers, that something doesn't get funded that should get funded. What about the danger is that something gets funded that should not get funded, in particular, because people are pledging because they think they're going to collect the bonus. And then that accident, and that ends up causing something to get funded. That actually wasn't worth it.

Yeah. So in equilibrium, again, that shouldn't happen. But of course, you're absolutely right. It could happen. I mean, public goods are provided, which shouldn't be provided all the time. Right. So we don't know, with our with our ordinary funding mechanisms, there could be forced riders, right, we have a forced rider problem, rather than a free rider problem. That is people are forced to pay for goods, which they themselves do not value. I think that that possibility is less likely here, though it is true, you know, even on today on Kickstarter, you know, people lie, or they're misinformed about the superduper coogler, and how well it's going to work and what can be accomplished, and so forth, you and so you can get public goods, you can't get campaigns being successful, which then end up failing later. Because the entrepreneurs are not capable of delivering and maybe reputation will help with that.

But the refund bonus makes it a more difficult, a bigger danger. Because let's say I think this I think this campaign is stupid, so it won't reach its thresholds. So I'm going to pledge to get the refund bonus. Yeah. And then a bunch of people, all of whom think it's it's going to fail, end up accidentally pushing it into success,

right? Yeah, no, you're right. Yeah. So I should, we should try and study that.

Yeah, I think there's actually anzar gram, I'm hoping that I'll pronounce this correctly, where some real world data from Agha Khan even though I thought that gitcoin was correct funding.

It is it's not about the dominant insurance contracts, it relates more to the extra incentives and the sort of inefficient allocations that come from it. But I think it relates to Mark's question. So the observation is that I'm in that system. For starters, we don't have hit this threshold or have a kick, you know, have the bonuses paid. But there is like a wide range of projects, grants, and they're ostensibly a theory in public goods. And what ends up happening though, is, you get a really weird modification to the allocation of resources in response to the perception of future airdrops, which someone commented about. And actually, this is an ongoing issue with like, I work with the gitcoin team. So we have the data. And we've been working on policymaking around what does and does not count at the platform level as an invariant theory and public good. Precisely because projects that have raised money or will raise money through Icos or VCs, etc. offer what is effectively kickbacks of various kinds, which influences the sort of action on Bitcoin in a weird thing about this is that it seems like they're actually using the gitcoin platform almost more to further their marketing, in the sense that it's highly visible. And it's a point of Central attention as the it's both crowdfunding platform for lots of open source projects, but also just like a kind of almost watering hole for this community are both from people who are very well endowed and people who are looking for funding. And I think it kind of speaks to this issue of efficiency of resource allocation, when there are other incentives composed and you can't really remove that, like in the real world, we don't get the lab setting we have to deal with in one form or another, either the market design directly or the governance of the platform, these like extra composed bits and what they do to the incentives and they kind of manipulate the equilibria in the games. And I thought it was particularly interesting with respect to Alex's work that he did talk a lot about, what I would normally understand is the preferential attachment dynamics, like the network is growing, people are seeing what other people are doing, and that affects the way people act. And that seems to be the dominant behavior we see in the gitcoin platform. So I think there's like some overlaps, but also some interesting stuff that you get from comparing field results to lab results.

Lovely. We had david Friedman, ping me a few times, David, Tevin. You are muted.

You very, very good. Yeah. I'm not really familiar with the experimental stuff. I'm a little bothered by the fact that Kickstarter in the equivalent I Really producing public goods. Because if you don't agree to contribute, eventually you don't get the good, which is not true of a public good. But what strikes me is the comparison between this and the older solution to the same problem, which is unanimous contract, that the unanimous contract solution, you say, you have a contract, you've got a list of all the people who benefit for the public good, you have a price for each of them, which is less than you believe the value of the public good is to him. And again, in the perfect information case, the only equilibrium is that everybody signs. And the reason that doesn't, one reason it doesn't work is that you don't have perfect information, you price it too high for one person, because you overestimate the value of the public good to him. And the result is that he refuses and the whole thing falls through. And as far as I can tell, in switching to the dominant assurance, in, in the sort of the equivalent version where you where you really want everybody to do it, which is not what you're doing in your experiments, you are slightly increasing the incentive to sign only in the sense that if you believe someone else is not going to sign, it now pays you to sign. Whereas before, if you believe someone else is not going to sign, you don't care whether you sign or not. But the cost of that you have to charge a higher price. Because the when it breaks down, you're gonna have to pay some money. So it's not clear to me at least at that level, that it's really an improvement over the other solution, they seem to have the same basic problem. Now also, both of them have a different problem, which I'm not sure you've you discussed. And that's the strategic problem, the person who says, If I hold out and refuse to sign this time, and make it clear why I can persuade the entrepreneur to do a second unanimous contract with my name left off, and that way I get the good for free. And I think you would probably have, I haven't thought about it enough, but equivalents to that problem in your in your version as well.

Yeah, so that's definitely true with the the dominant version where it's literally a dominant strategy. And that's because of the unanimity, as you were suggesting. But you don't need that. So even without requiring that, everybody, so you just have some contribution threshold, but you don't literally require that everyone could you be, then the only equilibria are the ones in which the public good is provided. So you're still right. You're also right, that this costs something because at least in practice, sometimes you have to pay the refund bonuses. And then the question is, are the successes worthwhile? If the entrepreneur makes some money on the successes, is that enough to pay for the refund bonuses? And does the entrepreneur do better is their profit higher paying the refund bonuses, and our results suggest that it increases the success rate by about 50%, which is enough, so that with a modest profit on the successful cases, the entrepreneurs are better off offering refund bonuses than not offering refund bonuses. That's

taking account of the fact that when you're offering refund bonuses, you have to be charging a higher price or not, you're just saying higher probability is enough to do it without a higher price.

No, no. So you'll have to charge. Yeah, you. Yeah, you have to change the contribution threshold so that the profit that you make is enough above your costs. So yes, correct.

So that that that is going to tend to cause more breakdown,

tend to cause more

rate raising the threshold is going to cause cause more failure. Because it may, it may turn out that at the threshold after the contribution, now, the individuals in your system to the individuals have to contribute a fixed amount or do they can create any amount they want.

They can create any amount they want. That has to be in some of the versions where there's a refund bonuses have a fixed amount, rather than proportional, they have to can contribute at least so much, at least a particular amount. So we avoid kind of contributing a penny in order to get a refund bonus of you know, $1. But in the proportional case, you can literally contribute however much you want. Remember the threshold, there's a cost. The entrepreneur has some cost of producing the public good. And there's a threshold how many people how much funding he requires in order to say he's gonna produce the public good. And if the if it was zero if the cost was if the entrepreneur could could produce a zero, then he would just set the threshold equal to the cost.

What I don't understand I'm sorry. Again, this is a real public good. So if I don't contribute, I still get it if other people do Yeah, yeah.

Maybe we take that into the chat. Okay. And I want to know from you, Alex, in the last maybe like two minutes or so what do you think is the most exciting near term use case for this? I know that you said a little bit about Kickstarter and like, you know, using it that maybe as an experimental ground, but like, which public goods? I think even in one of the papers, I think on libertarian, social engineering, you said, maybe even bridges one day, yeah, that way, but like, what do you think is like the pipeline? You know, how do you roll this out? Because then we're moving into like, the section where there's already a few projects that are trying to make headway on that. So I'm super curious, what do you think is the most high high level use case? And what do you think is the most short term?

Yeah, so I'm not, I'm not sure I want to broadcast the idea as widely as possible in order to get people to try it out. local public goods, I think make a lot of sense, to me, improving your neighborhood, things of that kind of nature. But there's obviously a lot of potential with smart contracts to produce public goods for platforms, right? I mean, that is a big problem, which governance of Bitcoin and aetherium and all these kinds of platforms, right. So on the one hand, you want decentralization, on the other hand, you want some governance, and you want to be able to improve the platform over time. And to do that you need some kind of funding mechanism, and you have a whole bunch of choices, you know, should we do? Should we double the block size? You know, or not? Right? Or should we double the block size? Or should we, you know, do X, Y, or Z, and this might be a way of getting people to contribute to those kinds of public goods?

Okay, lovely. Well, thanks a lot. And I think that's, you know, guess what, get confronted with quadratic funding, also. But thank you. Thank you very, very much, Alex, this was a lot of theory compressed really nicely. And I'm super excited about the Kickstarter projects, actually. And I think you meant they were mentioned in one of the papers, and I'm super keen to see them five. So that's awesome. Thank you very, very much. And we have now in the final, last 10 minutes, actually, two projects that are seeking to add to have some, some seeking to use some version of assurance contracts. One is in a non anonymous assurance contract. And the other one is actually dominant assurance contract, to actually Phantom stuff in the real world. So we have one project proposal from engineer a year on using anonymous assurance contracts and a platform called critical number. And I'm super happy to have him present here. I will share a background of where we talk about that in in the book and Anthony, please take it away.

Okay, thanks. That's great. So oops, started first one. This is a really interesting meeting. Thanks for the invite to talk. So we've heard a lot about funding, but they're also public goods, I think at some level that don't have to do with funding, but just have to do with people cording to get something done. So there's some coordination efforts where the only action really needed is not monetary, or even doing something but the Association of people's identity with some proposition. So some examples are like open letters by prominent individuals calling for something statement of intentions or a coordinated intentions to do something like a vote to unionize whistleblowing unethical or illegal actions, defection from leadership in a political party, for example, or pledges not even, you know, the necessary money, but pledges by high net worth individuals to fund a project with the pledges sort of good enough, you don't even necessarily need the money, like in the bank account. And these are all examples where and there are many other cases where there's a risk of financial or reputational or employment or political or even legal risk, in acting alone, with a disclosed, disclosed identity, but in which if people act in a coordinated way, and there are lots of people on board with the same thing that will be safe for all of them. So there's a coordination problem of how to get lots of people acting in a coordinated way when it's costly for any individual one to do it. So a potential solution to this, I think, is,

oops,

is to have people sign on in a way that their identities and their association between their identities and a given proposition is cryptographically secure, until a critical number that's where the name came from, or some other critical or some other triggering condition is met. And then all of those identities are made public. So the basic scheme for something like this goes as follows. So there'd be a set of propositions requirements, identity line, and a triggering condition or triggering conditions. And these are all set, and the propositions are recorded. And the requirements will actually I guess all of these are recorded on a public blockchain that's just for fun. Individuals with a verified identity who meet the requirements can then sign on to a proposition that they meet the requirements for whether an individual has signed a proposition is then kept cryptographically secure until the triggering conditions are met. So it's really inaccessible until the critical number has been met the people that have signed on, and then at this time the identities are made public. And if the triggering conditions are not been met by the deadline, all the data goes away and is just erased. So this is a way to basically ensure that everybody stays anonymous until everybody is safe to become publicly associated with that proposition. So there are a bunch of notes, I think, and implementing this. One is that the identification and verification has to be strict if it's a high risk proposition, or for you know, high, high, important figures, or if they're small numbers, because the major weakness of this mode is for, you know, if people want to out the group of people, to sign up lots of people to a proposition, maybe even fake people, or people that nobody cares whether they sign it or not, so as to trigger the triggering condition. So that's a, so verifying identity is a major issue with this, because it's sort of all about identity at some level, the requirements if if you're requiring that like, you know, 10 Nobel Prize winners sign on to this thing for it to go public, the requirements have to be verifiable at the same level as the security as identities. The triggering conditions have to be unambiguous, you don't want to weirdness in that the system has to be highly secure, preferably even if the whole system is compromised. So So it'd be nice if there was some necessity sort of for sort of interrogating the the signers before it really gets released. But it's hard to set up a system like that. I would say I can see that sort of centralized, but it could be centralized. But I think it's it's really important that it be simple to use, you know, there are a lot of I think, awesome ideas on the blockchain that haven't gone as far as they could, because they're not actually easy to use for people to to get involved. I think there are some other interesting variations or propositions could be legal documents. So actually, legal insurance contracts. And since you need a secure signature to to sign on to them anyway, you've actually legally signed this document, and it could just go into effect when enough people have signed it. And in the context of this conversation, I think these could be made dominant, though, I think it's less straightforward than in funding mechanisms, because you have to figure out like, where would the money come from? Who would be paying it? And so I think there are lots of interesting questions there. But I think you could have some kind of bonus, there is a free rider problem here, I think, in that you can just wait for other people to do the signing. The current status of this, I would say is that recording on a blockchain is easy, good cryptographic schemes exist. They were developed for project listo, not probably better ones could be done. Identity Verification is only partly solved. I would say creating an easy to use platform and driving widespread adoption is sort of a problem that any startup has. It's hard but doable. This is potentially a founding project of the moonlight Institute and a couple people here for that. But funding has not been secured for it. So it is currently exists at the stage of a white paper that I'm happy to provide for people who are interested on request.

Domestic Well, that was rather quick.

Five minutes.

And I'm hoping that yeah, I mean, if you want to request more information from engineer or even get in the chat for those from the moon and project who was still here, and then perhaps we can do that, too. And I think Mark Miller had a question. If you still hear back,

yeah. So on the whistleblowing case seems particularly problematic. What, first of all, to pursue to even propose such a arrangement has all of already carries all the risks of trying to anonymously leak the thing you're trying to whistle blow, and furthermore, setting up their requirements seems impossible. Because what would you would have to state some kind of checkable criteria, as to who else is a person credible whistle whistle blower for the same proposition, I just don't see how we couldn't work for a whistleblower?

Yeah, I think it's difficult. But I think as someone has pointed out there, the sexual harassment claims are sort of similar to whistleblowing in the sense that the problems are already already there. And this is a mechanism that has been tried in that context. That's what project Calista was doing. So I agree that it's it's challenging, but if you if the if the whistleblowing is essentially similar to, you know, in the sense that there are a set of people who are accusing a company of doing some particular thing, you could have a system that has matching claims of what that company is doing to each other, even though they're submitted independently, and only revealing things when enough independent, credible people making the same claim are matched up. But yeah, a lot of discussion there. I think that is one of the harder cases, though I agree.

And in the I think by paper, at least, I think engineer's taking it step by step going from low risk to higher risk ones. And that's definitely I think everybody everybody I would have been even about will be will be useful one if possible. And I think I mean, I'm gonna stay on longer if people want to discuss longer, but I think just in the interest of time, we also want to hear from Donnie in case some people, some of you have to hop off once we do have to hop off. And so Donnie, what who assurance contract product? Have you been cooking up?

Right?

Yes,

I'm, I'm calling my project good x, because I want to maximize the good via public goods. I'm very motivated by existential risk. I have been for a long time. And I think one of the core bottlenecks of many of the existential risks are human coordination. So that's why I'm interested in this space. It seems like the domino insurance contract is a good way to do that. Um, so yeah, I have a, I've whipped up a very minimal MVP. If you guys want to look at it. It's called conditional commitment. COMM. I'm very open to suggestions and direction. I'm kind of thinking about a Kickstarter for network states as like a good direction to go in. That seems very ambitious. And like, if you can create a new country with a dominant trends contract or something, then like, that's pretty like public goods complete. So but I'm very, like, I'm very interested in like, Can we get groups of people like moving to the same location? So that's even like, from smallest to bigger scale, like attending the same meetup or group house, or village or charter city? Um, so that's the public that though is like how, how to verify if you're doing things besides crowdfunding, like, how do you how do you verify that people say their gun? You know? Yeah, 10 of protest? Or like, stop eating meat for a month, or whatever? They're coordinating on? How to cover for that? Um, I don't know. So, yeah, um, any questions?

Thank you. Fantastic. I just want to say for those of you who have to hop off, feel free to do so in case LS engineer or Donnie have to do so I will say on fallen longer for those who still want to, and just in case they are not friends, I know that I will share Anthony's and Johnny's and Alex follow up. If If you guys send it to me and lsps, your question your slides to they were fantastic. But yeah, and I think we could move to a q&a as well. I loved how Tony, how you said, group houses or charter cities, one or the other, whatever. Very, very similar. But Okay, any questions? Oh, yeah. Well, I would love to discuss between Anthony and Alex, maybe. And I think I already said this in the chat. But, you know, could there be a way in which one could make anonymous issuance contract also dominant? Is there something to be said for having something like a whistleblower, kind of like almost bound in by which, you know, you could even be more incentivized to go first. And you'd get the refund paid if, if nobody else was brave enough to actually put the named offers, even though that would be very, very hard to do, especially with a danger that mark pointed out, and by verifying who even does that and who pays the price, and so it could also increase increase the dangers of it, but I would love to have your opinion on that Alex and entity.

I think one difference with what Anthony is is talking about, which I think is very interesting, is that when Anthony is talking about you're accumulating information over potentially a long period of time, and there's no there's no point at which you would say okay, this is the cut off and we're going to give the refund bonuses. Like if Sexual harassment calisto kind of setup, you just really want to keep that open. So there wouldn't be a point at which you would say, Okay, now we failed the contract. We didn't we didn't meet the traffic, we didn't meet the threshold. So we're going to give the refund bonuses. If you're contribute, if you're if you want to be cumulative, then you just might say, Well, we've just haven't reached the threshold yet.

Yeah, I think you could run it that way. But I think you could, you could run them also with a, with a deadline, and with a, you know, publicly visible ticker, you know, for how many people have signed on. So it could be that, you know, something gets paid out, if you if it fails, that that payments get paid out. And then they you know, all the information gets erased, you'd have to then, you know, have some way of knowing who to pay, you know, that would have to be part of the identity verification system, which makes it a little bit trickier. I mean, there's what's very different about it, is just, you know, what you're, you're bartering is your is reputational, or, or identity or position rather than money. You know, money is very fungible, and those other things are not. So I think that leads to very different dynamics. But in principle, I don't see why there couldn't be a way to reward reward on failure in the dominant method.

Thanks. And then I guess, you know, one, when I whenever I think of use cases, like one use case that comes up to me would be something like an insurance contract on reaching escape velocity and aging. And I know that this sounds a little bit like, you know, very future, very future fetched. But just because I think there's many people that would potentially donate much, much more to aging research, if they knew that there was some threshold that would make it much more likely for them actually to reach escape velocity. And there is no such thing. And I've talked to a few a few folks who said, well, that number will be insanely high. Well, yes, it may be. But there's also maybe an insanely, insanely higher number that we don't know of people who would be willing to donate if there was such a thing. So maybe you want to create something like the prediction market on which mechanisms are especially relevant for escape velocity, how much money one would need to put in each of those areas for them, at least to have, where the prediction market would have some would have some, some, yeah, some assurance that that part will get solved, and then only create an assurance contract for those bids, if another if enough others also donate for that particular aging milestone to be solved. And so I think that that may actually unlock a bigger funding pool than we currently have. So that's my kind of like, very, very high pie in the sky dream idea. I would love to know a What do you think about that? Or do you think that there's a particular cause? I mean, Alex, you already talked to this, but like where you would really like to which he would be react to some type tackled with your proposal?

Well, as one of the 7 billion people who will probably eventually die, I'm all for favor. Call in favors. So, so yeah,

just make people pay half their net worth when they reach 128? One?

Yeah. Yeah. Yeah.

That'll get you a lot of money if you're successful.

Yeah. Good one.

And enter new, Donnie, I know that you talked a little to use cases. But is there a particular one? Where are you like, if only we could do that? That would be tremendously helpful as a as us because there are so many ambiguity? Yeah,

I see it as Yeah, there are a lot of use cases in my one of my goals is just to to hopefully, foment bring into being a very general purpose platform for doing this sort of like meticulous in the sense that there are lots of use cases, and you can handle them with one thing. I do feel a little bit of something interesting with funding, I think, is that when you start to talk about really large amounts of funding, like you know, might be required for longevity, and really, really wealthy people who might be able to provide it. I think it's the the agreement to provide that funding, potentially in writing is almost as valuable as having that funding sort of committed in the bank or like an escrow or something, but it's much more likely, you're much more likely to get it like I think most billionaires are not going to put like literally a billion dollars in escrow somewhere. They don't want to do that. But they might commit to, you know, in writing, if this is if their 999 other billionaires will do this with me, I will give a billion dollars and that might be good enough to get like eight or 900 of them to actually give the billion dollars at the end of that might be good enough. So I think there there are potential realities for really high funding levels that might actually be better suited to something like a, an insurance contract like this, rather than actually, you know, Kickstarter thing.

Well, that would ideally combine. Maybe something like a prediction market, like a meticulous with a split contract, which would be half automated and a half and pros maybe with with a dominant shows correct. Not anonymous yet though, so we don't hit all the buzzwords. But um, Okay, last question on my end at least what can people do to help with? With your projects? Like what would be really useful? I know that you already shared a little bit but like maybe Alex, engineering, Johnny, just that we haven't in one go like, what are all What do you want people in the industry to do to help push this forward? Like what's kind of like an actionable item?

For my part, I would say, you know, anyone who is or know someone, well, who's really, really expert on the cryptography side, some good conversations with them would be incredibly helpful. So you know, get in touch, I'm happy to send you the white paper just or set up a meeting. I'm not a cryptography expert at all. So I'm pretty far afield on this. And I think it has to be done really well.

on my end, if you're a programmer and wondering what to work with me on this, that'd be great. That's good. I'm pretty good on funding, but you know, that, that would help getting a programmer if they want money. But ideally, I'm also interested in back certificates as a project. So maybe I just hang out, in fact, certificates,

opening up a whole nother box of Pandora.

on your end.

Well, our next goal is to do a field experiment. So yeah, if you have any ends at any of the platforms, like Kickstarter, we talked with the guys that Kickstarter, they're very, very reluctant. But we may go back to them, but there if there are others, you know, GoFundMe or something like that, who want to try this out, let us put us in touch.

Okay, well, one wonders why they're reluctant to that, I think. And, okay, well, I know that I mean, I don't know if you have to vote. I know that there's two more hands raised here and feel free if you have to go and drop off any time. Otherwise, I'll just keep on taking hands. Move on. Until one of you drops off some of our bread timboon had one.

Yeah, no answer. I was interested to hear stuff. I've been proposing systems like this for about 15 years. And what I found, though, is that it's actually pretty hard to describe them or to sorry to implement them. cryptographically. Although the creation of blockchains may have changed that I haven't really researched what could happen. So I've ended up mostly designing systems that use non cryptographic taking students now. I mean, you mentioned Kalista, which does it for sexual harassment reports, by just using a central trusted authority. But like one of the problems I thought would be really great to solve would be reporting of corruption. And one of the problems with reporting corruption is that if it's disclosed that you reported corruption, like of a mobster, this could have very negative consequences for you. And so the monster can easily threatened a bunch of people break their legs if they don't submit false reports on him, just so he can see who all the people who reported on him are. So one simple solution was it when you go to the ballot box, which is this sort of secured anonymous space for communication, that you could also write down in your ballot box, the most corrupt official in my city is, and you would just fill in that name. And then the results will be published next day. In the end, if you're the mayor, you don't want to win that particular election. So non cryptographic systems seem to be much also easier for people to understand and easier for people to believe the other system I was proposing for the impeachment vote was a system where everyone would vote with a vote inside envelopes. And if you didn't get 51 envelopes, you burned the envelopes in front of everybody. So you never looked inside them to see the names of the people who voted. And these sort of techniques are actually much clearer to people that they can believe in the security. One of the difficulties with cryptographic techniques is getting ordinary people who must participate in these to feel that they're truly safe in you know, knocking out a mobster, because you got to refill you're really safe before you knock out a monster.

Yeah, yeah, I would say I just agree with all of that. I think those are great points. My view is that the the best path is to make something that works well for like less important lower security things and it's easy to use, and then to sort of steadily try to build trust from there. I think if you try to go initially for like, mobsters and whistleblowers for major corporations or defense leaks and things like that, you're you're taking on the very hardest cases, you know, to start and that's that's a difficult way to go. So I I would advocate, you know, creating something that can be used in in simple cases, like people who are signing a petition, where it's not really that important, if they're if their identity got leaked. And then sort of tried to build up the trust, the trust and the adoption, and you know, people being signed on to it. From there. I think it's also I mean, what is very helpful is if you have, you can have something where people kind of have their identity verified and get onto the system for very low important things. But then they can use that secure identity and higher importance things later. Because you don't want people signing up, sort of just to do one thing, or else. That's,

and that's certainly what will happen. I mean, you know, not people will not voluntarily sign up, I want to be in the sexual harassment reporting system while some people would, but a lot of people wouldn't until after the fact. But even signing up for that could be an issue even I'm in the petition signing system is something you'd probably not do until you want to sign your first petition, which would eliminate that ability. Now, if you're talking about votes in the Senate, getting 100 senators to be issued tokens, which let them make cryptographically secure votes. That's a doable thing. But for larger populations, it's hard. Yeah.

All right. And maybe that's one Mike. Mike, you don't have a video, so I'm not sure if you're, if you realize that you're on mute. Okay,

okay. Can you hear me? No? Yeah. Okay. Oh, yeah. Question for Anthony, in particular, a similar line. So Brad just asked, and that's as far as Have you found a way to contract contracts on the blockchain? Right.

Yeah. Okay. microphone,

your audio is really poor.

Can you hear me better?

How about that? Much better?

Okay. any better?

No, that's great. This is great.

Oh, okay, good. Yeah, I had a question as far as who's Anthony come up with a way to encrypt data on the blockchain, and then trust lessly provided key if the conditions are met to be cryptid. I ran into this problem with a similar type of app a few years ago. And then there was no way known really to encrypt or hide a contract on a public blockchain without being publicly accessible or encrypted data, you know, and decrypted trust lessly.

Right. So and so my, the the place that I see for the blockchain, and this is encoding the statements and the condition so that everybody agrees that you haven't changed those, like retroactively. So I think there should be a public record of that. That's unimpeachable. Doing the doing the secure matching of the participants with the propositions could maybe be done on the blockchain. But I'm not an expert in that and have not, I envisioned that more as a central authority, who would who would hold that information? But I'm not even sure exactly how the best way to do that is I think there are, it seems like there are good ways, but I'm not expert enough to know how good they are, or how much better they can be made.

And it would have been able to join but, you know, I mean, like not just to say, oh, zero knowledge proof will solve it. But, you know, if there was a way in which, you know, there were a few certain criteria that one would have to fulfill to, let's say, be the relevant whistleblower. And one could maybe prove that one has access to that information via as your knowledge proof, in which case, you wouldn't even need another, another centralized entity that creates those signatures and have like this kind of leakage of trust to this third entity. So maybe if you could, you know, set a few specific conditions that one could only know if one proved a specific thing was even old school, then that may be one way in which one could you know, verify that one is the right kind of whistleblower that one actually wants and not, let's say, a number of agents from the US or Chinese government that is trying to trying to, you know, coax out a few of the laws without providing them safety numbers. So that was my initial idea. But Zuko would probably scope me now because I said something wrong. But yeah, if people come up with really smart schemes to do this, that would be really useful, I think, even just you know, on kind of like low stakes use cases. And one would have to then level it up.

I've considered it a delegated proof of stake and the delegates can hold like a sharded set kind of like Canavan, multi SIG that if the conditions are met, then they agree vote on to release key if a majority agree, so there's some ways out of to do it. That's what I have to say anyone else has come up with any solutions for it. Okay, thanks.

Well, we need experiments clearly. Yep. Yeah, it was lovely to see all of you. Oh, James, did you want to say something?

I was just saying Mike the only the only folks I don't know how they're doing it but but I do know folks on on bsv Bitcoin SV blockchain are doing a built encrypted chat over like we're all on chain. And I know a bunch of those folks swell, go ask them how they're doing it. I don't actually know if they're doing what you think on what you'd like to be doing. But they are doing encrypted chat over, like all on chain. So, okay, crypto between users. between users? Yes. Oh, great. Okay, that's a very similar problem that not to look into that. Well.

I think z cash is based on eBay, which is, you know, trying to do an eBay as your knowledge base, eBay. And they are starting with a social media platform first, which is also using the knowledge base for encrypted messaging. And then, you know, eventually, one could even share one's address. And as your knowledge base, I think, you know, they could also be interesting people to talk to, I know that they are collaborating closely with the Keshe Foundation. So I'm trying to get them on to our next privacy discussion with Zuko. So maybe, you know, we can actually talk about a few privates we're throwing ways to do this kind of thing. At our next meeting there, which I'm still scheduling. Okay, anything else that we're missing? No. Thanks, Alison.

Well, thank you so much. Thank you so much for staying longer. It was really, really lovely Alex, and he totally loved your presentation. And please send me the slides. And I need to go to the book now and edit a lot because there was okay, everyone, I am going to send out a new meeting invitation for a meeting next week with a chip claros like a privatized, like arbitration service, and maybe even legalese, but I'll send that out for next week. And yeah, we're gonna keep on keep on writing on a weekly basis. Now for those who want to thank you, everyone. Bye.