We don't have money but we can tell you how to get it kind of or at least how to do it safely. All right. So the basic idea of this talk uh and I'll go through it a little bit more uh densely later is to build a smart contract that is strategically equivalent in a kind of like a formal and rigorous way with a Y combinator safe. So just once again building a smart contract that is equivalent in some sort of like sense to what a safe legal contract is. If you don't know what a safe is it is the kind of a very generic way that is very common in the VC industry for a startup to take equity or to take investment not exactly equity.
All right uh outline of this talk I'll go over what the problem is. Uh we'll talk about clause the game which is kind of the prerequisite project to the safe for Dows project. I'll give a little bit of intro for the nerds out there to compositional game theory which is the mathematical framework that underlies all this. Uh Lance will talk about the YC safe and why it's so awesome and why so many people use it. Uh he'll also go into the safe for Dallas smart contracts what we've already been building as well as a sign of future work and how you all if you're interested can get involved.
All right so really quickly what is the problem? Well, Dows need to raise money. Uh, and the standard way to do so on chain is to have a token. But just like other startups, uh, you know, Dows, many early Dows, like it's, you really don't want to start a token in a lot of cases because it's kind of hard. There's all legal risk.
It's just a extra complication on top of whatever product you need to build. Uh, it's yeah, it's not 2017 anymore. So, the alternative is to take money through some sort of legal entity. But many Dows don't want to have entities. uh they want to be entityless and even those with entities you know simple agreements for future tokens uh which is the kind of the SAFS if you're familiar with them they're legal contracts but they're not standardized and generally offer like very poor investor protections uh investors don't like them they are not equivalent to safes what's the idea here once again it's to build a smart contract that is strategically equivalent to a yometer safe that takes all the work that's been done to understand what the hell like what a safe is, why it's a good institutional kind of setup or game that investors and startups want to play and duplicate that and bring that on to a smart contract so people can you know have that safety and security in making investments to a Dow that's building something.
All right. Uh so basically how we're building this is we start with something called clause the game which is a software framework we built that uses compositional game theory to build game theoric models of whole legal contracts. So going won't go too deep into like what exactly that means the point is we use the software framework to construct a game theoric model of the safe legal contract. Uh and then we use this model to specify kind of in a rigorous way a smart contract that actually duplicates the game that's being played whenever you know a bunch of people you and an investor sign a safe. And the background uh to this is this is being done in the context of Dowstar which is a nonprofit 100% you know public good um standards body for Dows and this is uh also interacting with one of the other working groups in Dowar was called Millennium Falcon that focuses on regulatory interoperability for web 3 basically how we can better interact with regulators.
All right uh so some plastic game examples just going really quickly. Yeah. So before hitting Y combinator safe contract, we did the same thing for generic sales of good contract with a kind of a legal startup. We've also built models for uh carbon credit carbon credit markets, water management systems in Nepal. It's actually really fascinating if you're into this stuff as well as kind of landlord tenant kind of mechanism design.
Okay, a little bit background on compositional game theory. I will take just five seconds. It is a mathematical framework. It's based on something called category theory. If anybody of you are like super nerds, you might have heard of that.
It's the most awesome piece of math ever. I'll just end it there. Uh if you want to dig in, just come up with me afterwards and I'll talk about it. It's based on these things called open games. Uh and what an open game is is something that looks a little bit like that, if you're familiar with kind of control flow.
X and Y is kind of input and output. It's like the observations you have and then the play you make. But this is once again, this is a model of strategic behavior. and strategic in strategic behavior you need to sort of think about like what's stuff that's happening in the future. So we have like arrows going backwards in time.
So R and S that kind of represent utility and sort of feedback signals. So this is kind of the underlying sort of mathematics uh behind this work. Uh once you have that sort of really simple mechanism, you have lots of different ways of sticking together and building more complicated games. parallel composition, you stick these things in parallel simultaneously. Sequ uh sequential composition in which you're sticking them for one game and then another game.
And uh you can stick all these things together to get you know generic things like prisoners dilemma uh where you have two players playing a simultaneous game. Why there uh y 2 here which are the plays and they're getting a reward from some outcome function u uh back to them. All right, here I will hand it off to Lance. Cool. So, we w we talked about compositional game theory, but we think of compositional game theory as like programming languages for game theory.
So, the compositional game theory engine is an engine that's written in Haskell that is used to actually program these compositional games. And like we talked about with clause the game, you can program these compositional games to kind of fit together and draw an EP uh a larger economic model for something like a contract, right? So we have the YC safe. Uh I just want to see a show of hands. Who's familiar with a safe contract?
Okay, awesome. Most of the crowd. Uh just for everybody, a safe contract is a way for startups to raise finances without selling equity because there's a lot of regulation that's attached to equity. So YC created this safe for uh startups. I I know if you can see a little bit of the verbiage.
It's a little bit legal, but uh we break that down in the actual uh game theoretic smart uh game theoretic model and then the smart contracts as well. So this is the model that we created for the safe. So as you can see uh in the last couple slides. So this is just a simple game prisoners dilemma. But the game for a safe smart contract is a little bit more complex.
But walking through it you have an investor who agrees to give money to the company which is the safe agreement. And the company at some later time they decide to raise money right? So they have an equity financing round. they have a series A that's modeled with this game right here. Other investors come in, they invest, and when the safe um when when the equity financing round happens, the safe investor shares, they convert.
So that you start off with just a percentage of the company, but then after the equity financing round, your percentage converts to actual stocks that you can use to either sell. At this point, the safe investors can say, "Hey, I want to sell my stocks or I want to keep them for long term." That's where this option is. And then um after afterwards, they'll go through a um another decision to reinvest in a future equity financing round. And in a future equity financing round, then maybe like series B, series C, they can re-enter subsequent rounds.
But we've modeled this game theoretically from start from the first safe contract all the way to an exit which is um an IPO a merger or an acquisition uh or if the business is unsuccessful a dissolution. So, so after we created the um actual compositional games in Haskell, we decided that we wanted to build something that people can actually use uh that Dows can actually use to raise money on chain. So, our smart contracts are made um they're based off the safe compositional game that you just saw and they're an ERC 1167 proxy contract pattern. So if you're a DAO or you know people who are leading in a DAO, you can basically get a copy of this smart contract and use it to raise funds um or to raise a safe. And the thing that we did is we kind of handled the logic all the way up to an exit.
And these are not yet deployed on testn net still working through bugs but you know actively looking for contributors if you're interested in that. So just a quick overview of the smart contracts, right? We have the actual safe cap table. You'll notice that we have uh an array of investors. We have a mapping.
So each shareholder, they have a certain percentage of the company. And then the uh total amount invested. So that's the total amount the startup has raised with safes. And then you have the valuation, right? So the company has a certain valuation that investors in the equity financing round are going to determine if they want to invest or not.
But then you have another strct in the smart contracts that is a capitalization table. Now this is different from the safe cap table because in a safe remember you're just selling a percentage of the company but you're not actually selling shares. It's a simple agreement for future equity not current equity. So this capitalization table is made for when those when uh that percentage converts into actual shares. They're distributed and you can see that there is a um price per share and then a total number of shares.
So you actually get the logic for uh for your shares and then you um if you're a creator you initialize the smart contract. So this basically goes through the logic of um a Dow uh a Dow manager. They create the smart contract. They initialize it with some funds. They end up being the manager of the proxy contract and then they're able to mint the amount of shares that they want.
So let's say I want 10 million shares in my company. Um then I can do that. And they're they're represented by ERC20 tokens. So your shares aren't something on a paper document. they're something that's actually onchain, a lot more secure.
And then your valuation is set to zero because typically when you start a company, you're starting at zero, but you can add in um initial funds if you raise like a friends and family round. So then we get to the step where investors can actually put money into the company. So you have uh an investor, they have a certain amount that they want to invest and then you have a post money valuation. So the post money valuation is really important in safes because you want to determine how much the company is worth and how much the owners how much the employees like options and how much the investors own. So you know of course you want to invest if you want to invest you got to invest more than zero dollars.
Um and then you have to actually be an approved investor. So the way it works in the smart contract is the founders will approve an investor and once the investor is approved then they're good to um to submit their investment to the smart contract. So after they submit their investment then we add them to the list of investors. we uh we calculate their the percentage of the company that they own and then we delegate their uh or we get we kind of rearrange the percentages of the company, right? So we take a little bit of equity out of the founders wallet um out of the founders wallet and then we give it to the new investors because that's how it works before you um raise raise money in the equity financing round.
So if you want to actually look at the smart contracts, we you can scan this QR code. This is the GitHub. I'll get out of the way. That's the GitHub to the safer dow smart contracts. And then this is the open games engine.
It's written in Haskell, but if you are really interested in economics and game theory, it's honestly an amazing way to get involved with um game theory, compositional game theory, uh and category theory. Everyone good? I'll give it a second. It's not over yet. Okay.
And so this is still a work in progress and we really we really want to encourage everybody to um you know at the end we're going to have our contact information, but reach out to us. This is a public goods service. So we really want Dows to be able to use this without having to worry about anything. you know, you handle your you don't have to hire in uh a CFO to handle your uh equity fin your equity financing or your fundraising. You can just do it automatically on chain.
And this is, you know, once again is a project with Dowstar. Uh we actually have a telegram for the safer dows project. So if you reach out to one of us at the end of the presentation, uh we'd be glad to talk to you about it. Um oh yeah, and then also if if you're a lawyer, that would be very helpful. uh particularly somebody interested in contracts, we need a lot of legal expertise and thank you very much.
You can contact Josh there and this is my uh link tree. Thank you so much. Thank you Lance and [Music]
Automatic transcript — names and jargon may be misspelled.