Loophole: Adversarial Agents To Stress Test Your Morality — Brendan Rappazzo, Morgan Stanley

I'll be talking about my project I'll be talking about my project loophole. And I'm actually a machine loophole. And I'm actually a machine loophole. And I'm actually a machine learning researcher at Morgan Stanley, learning researcher at Morgan Stanley, learning researcher at Morgan Stanley, but this has nothing to do with Morgan but this has nothing to do with Morgan but this has nothing to do with Morgan Stanley. This is just a open-source Stanley. This is just a open-source Stanley. This is just a open-source project I've been building for fun. And project I've been building for fun. And project I've been building for fun. And to give sort of the high level flavor to to give sort of the high level flavor to to give sort of the high level flavor to start, it's really this uh game you can start, it's really this uh game you can start, it's really this uh game you can play that's built on top of this play that's built on top of this play that's built on top of this adversarial agent framework. So you adversarial agent framework. So you adversarial agent framework. So you specify your morals, one agent codifies specify your morals, one agent codifies specify your morals, one agent codifies that into a legal system and then these that into a legal system and then these that into a legal system and then these two adversarial agents try to find two adversarial agents try to find two adversarial agents try to find contradictions in your morals. And contradictions in your morals. And contradictions in your morals. And lately I've been building different lately I've been building different lately I've been building different extensions on top. Um, but I wanted to, extensions on top. Um, but I wanted to, extensions on top. Um, but I wanted to, you know, start with sort of the origin you know, start with sort of the origin you know, start with sort of the origin story and and how I came up with this story and and how I came up with this story and and how I came up with this this idea. And so this really started, this idea. And so this really started, this idea. And so this really started, you know, a long time ago, I had sent my you know, a long time ago, I had sent my you know, a long time ago, I had sent my DNA into 23 and me for uh ancestry DNA into 23 and me for uh ancestry DNA into 23 and me for uh ancestry testing. Um, and I kept hearing about, testing. Um, and I kept hearing about, testing. Um, and I kept hearing about, you know, more recently how DNA samples you know, more recently how DNA samples you know, more recently how DNA samples can be used, of course, to help solve can be used, of course, to help solve can be used, of course, to help solve crimes and all these forensics and cold crimes and all these forensics and cold crimes and all these forensics and cold cases. And I was thinking about how I cases. And I was thinking about how I cases. And I was thinking about how I had sort of opted out of of everything had sort of opted out of of everything had sort of opted out of of everything because, you know, I was scared of the because, you know, I was scared of the because, you know, I was scared of the kind of slippery slope and and how my kind of slippery slope and and how my kind of slippery slope and and how my DNA would be used. Um, DNA would be used. Um, DNA would be used. Um, but you know, there there are certain but you know, there there are certain but you know, there there are certain cases that I would be okay with. And cases that I would be okay with. And cases that I would be okay with. And it's sort of interesting. I was thinking it's sort of interesting. I was thinking it's sort of interesting. I was thinking like, you know, if someone could present like, you know, if someone could present like, you know, if someone could present to me case by case, you know, we'll use to me case by case, you know, we'll use to me case by case, you know, we'll use your DNA to solve, you know, help solve your DNA to solve, you know, help solve your DNA to solve, you know, help solve this cold case or this murder. I could this cold case or this murder. I could this cold case or this murder. I could sort of say yes or no. and I know where sort of say yes or no. and I know where sort of say yes or no. and I know where the the definition of like the and the

the the definition of like the and the the the definition of like the and the nuance of my morals are. Um, and but you nuance of my morals are. Um, and but you nuance of my morals are. Um, and but you know, of course, like enumerating all of know, of course, like enumerating all of know, of course, like enumerating all of this case by case is really sort of this case by case is really sort of this case by case is really sort of cognitively prohibitive. Like there's cognitively prohibitive. Like there's cognitively prohibitive. Like there's not a good way to do this currently. not a good way to do this currently. not a good way to do this currently. And then I was thinking sort of more And then I was thinking sort of more And then I was thinking sort of more zoomed out that there's a lot of zoomed out that there's a lot of zoomed out that there's a lot of analogies to sort of the legal system as analogies to sort of the legal system as analogies to sort of the legal system as a whole. So, you know, one way to think a whole. So, you know, one way to think a whole. So, you know, one way to think of what a legal system is in our society of what a legal system is in our society of what a legal system is in our society is really just a way that we are trying is really just a way that we are trying is really just a way that we are trying to c codify our own moral beliefs. And I to c codify our own moral beliefs. And I to c codify our own moral beliefs. And I think sort of in a similar way like think sort of in a similar way like think sort of in a similar way like finding the true nuance of our morals finding the true nuance of our morals finding the true nuance of our morals and what the law should be as this and what the law should be as this and what the law should be as this really hard translation task. And I really hard translation task. And I really hard translation task. And I think often we kind of heir on the side think often we kind of heir on the side think often we kind of heir on the side of being too general because you know of being too general because you know of being too general because you know finding that nuance is really difficult finding that nuance is really difficult finding that nuance is really difficult and if you try to have a perfect and if you try to have a perfect and if you try to have a perfect translation it can we lead to these kind translation it can we lead to these kind translation it can we lead to these kind of weird corner cases or or weird of weird corner cases or or weird of weird corner cases or or weird failure modes and I even think kind of failure modes and I even think kind of failure modes and I even think kind of like peer-to-peer when we're relating to like peer-to-peer when we're relating to like peer-to-peer when we're relating to each other politically a lot of times each other politically a lot of times each other politically a lot of times the disagreements are more kind of the disagreements are more kind of the disagreements are more kind of fighting over core values which we don't fighting over core values which we don't fighting over core values which we don't really disagree on um instead of really disagree on um instead of really disagree on um instead of exploring the really nuanced exploring the really nuanced exploring the really nuanced points of our our morals and I think you points of our our morals and I think you points of our our morals and I think you know following that kind of broader know following that kind of broader know following that kind of broader legal example I think in the you know legal example I think in the you know legal example I think in the you know the English common law system that's why the English common law system that's why the English common law system that's why we sort of lean on case law so heavily we sort of lean on case law so heavily we sort of lean on case law so heavily because we know that finding this this because we know that finding this this because we know that finding this this nuance and these nuance boundaries is nuance and these nuance boundaries is nuance and these nuance boundaries is difficult and so we kind of rely on

difficult and so we kind of rely on difficult and so we kind of rely on smart judging to interpret and apply the smart judging to interpret and apply the smart judging to interpret and apply the law um correctly and you know of course law um correctly and you know of course law um correctly and you know of course even with like the Supreme Court things even with like the Supreme Court things even with like the Supreme Court things can get elevated and we can decide can get elevated and we can decide can get elevated and we can decide whether a law is valid at all. Um and so whether a law is valid at all. Um and so whether a law is valid at all. Um and so that was sort of you know the idea is that was sort of you know the idea is that was sort of you know the idea is like can you take your morals and can like can you take your morals and can like can you take your morals and can you do this kind of synthetic case law you do this kind of synthetic case law you do this kind of synthetic case law generation. So you know um this is generation. So you know um this is generation. So you know um this is overwhelming to do by hand but it seems overwhelming to do by hand but it seems overwhelming to do by hand but it seems like the new generation of LLMs are like the new generation of LLMs are like the new generation of LLMs are finally sort of smart enough to do this finally sort of smart enough to do this finally sort of smart enough to do this kind of highle moral reasoning and so kind of highle moral reasoning and so kind of highle moral reasoning and so that was sort of the the the starting that was sort of the the the starting that was sort of the the the starting point for this game. Um, and I just want point for this game. Um, and I just want point for this game. Um, and I just want to take you through sort of the initial to take you through sort of the initial to take you through sort of the initial release of the game, the setup, how it release of the game, the setup, how it release of the game, the setup, how it works, and then also talk about some of works, and then also talk about some of works, and then also talk about some of the different branches I've been the different branches I've been the different branches I've been building on top of this open source building on top of this open source building on top of this open source project because I think it could go in project because I think it could go in project because I think it could go in some interesting directions. And so at a some interesting directions. And so at a some interesting directions. And so at a high level, how the game works is you in high level, how the game works is you in high level, how the game works is you in natural language and this all happens. natural language and this all happens. natural language and this all happens. It's sort of like a terminal based game. It's sort of like a terminal based game. It's sort of like a terminal based game. You specify your morals and it could be You specify your morals and it could be You specify your morals and it could be uh your general morals or maybe about a uh your general morals or maybe about a uh your general morals or maybe about a specific subject and then there's one specific subject and then there's one specific subject and then there's one agent that takes those morals and drafts agent that takes those morals and drafts agent that takes those morals and drafts sort of a really rich legally codified sort of a really rich legally codified sort of a really rich legally codified legal system and then it just operates legal system and then it just operates legal system and then it just operates in this loop where one agent is in this loop where one agent is in this loop where one agent is instructed to try and find loopholes in instructed to try and find loopholes in instructed to try and find loopholes in your system. So something that is uh your system. So something that is uh your system. So something that is uh immoral but legal and another is uh immoral but legal and another is uh immoral but legal and another is uh prompted to find overreach. So things prompted to find overreach. So things prompted to find overreach. So things that are actually moral but illegal

that are actually moral but illegal that are actually moral but illegal given your system. And a judging agent given your system. And a judging agent given your system. And a judging agent looks at the your morals the produced looks at the your morals the produced looks at the your morals the produced legal code and these sort of synthetic legal code and these sort of synthetic legal code and these sort of synthetic case law examples and first sees can it case law examples and first sees can it case law examples and first sees can it auto patch. So, like maybe the original auto patch. So, like maybe the original auto patch. So, like maybe the original draft of your your legal code sort of draft of your your legal code sort of draft of your your legal code sort of was an imperfect translation and there's was an imperfect translation and there's was an imperfect translation and there's not really a contradiction and it can not really a contradiction and it can not really a contradiction and it can just sort of autodo this update. Or just sort of autodo this update. Or just sort of autodo this update. Or maybe it's really kind of an maybe it's really kind of an maybe it's really kind of an underspecification of your morals um or underspecification of your morals um or underspecification of your morals um or some kind of contradiction in your some kind of contradiction in your some kind of contradiction in your morals and in that case it raises it to morals and in that case it raises it to morals and in that case it raises it to you as the user to sort of be the judge you as the user to sort of be the judge you as the user to sort of be the judge and make a determination. and make a determination. and make a determination. Uh Uh Uh >> is it still on for you? It disappeared >> is it still on for you? It disappeared >> is it still on for you? It disappeared for me.

Okay. Um, so I know that, you know, if Okay. Um, so I know that, you know, if you I hope if you're curious about the you I hope if you're curious about the you I hope if you're curious about the game, you'll play it. It's all on game, you'll play it. It's all on game, you'll play it. It's all on GitHub. But I just wanted to show some GitHub. But I just wanted to show some GitHub. But I just wanted to show some examples. And this is a lot of text. So examples. And this is a lot of text. So examples. And this is a lot of text. So it's more about just showing the kind of it's more about just showing the kind of it's more about just showing the kind of shape of the the input and output. So shape of the the input and output. So shape of the the input and output. So this is sort of how you would provide this is sort of how you would provide this is sort of how you would provide your input. And going back to the DNA your input. And going back to the DNA your input. And going back to the DNA example, you might you know specify some example, you might you know specify some example, you might you know specify some number of of moral principles. And then number of of moral principles. And then number of of moral principles. And then the sort of codified legal system again the sort of codified legal system again the sort of codified legal system again just kind of looking at the shape has just kind of looking at the shape has just kind of looking at the shape has this really like legal ease, you know, this really like legal ease, you know, this really like legal ease, you know, preamble articles uh sections really preamble articles uh sections really preamble articles uh sections really trying to be uh you know write it in trying to be uh you know write it in trying to be uh you know write it in precise legal language. And then these precise legal language. And then these precise legal language. And then these the different kind of synthetic case the different kind of synthetic case the different kind of synthetic case laws get suggested. So in this case it's laws get suggested. So in this case it's laws get suggested. So in this case it's talking about this is a loophole it talking about this is a loophole it talking about this is a loophole it found where an insurance company um found where an insurance company um found where an insurance company um trained a predicted machine learning trained a predicted machine learning trained a predicted machine learning model not on your DNA but on artifacts model not on your DNA but on artifacts model not on your DNA but on artifacts of the DNA. And so it's saying you know of the DNA. And so it's saying you know of the DNA. And so it's saying you know this is actually immoral but currently this is actually immoral but currently this is actually immoral but currently legal given your system. And in this legal given your system. And in this legal given your system. And in this case, it's it found that it could do case, it's it found that it could do case, it's it found that it could do sort of this auto patching and then you sort of this auto patching and then you sort of this auto patching and then you get this sort of like get style get this sort of like get style get this sort of like get style difference of of your original legal difference of of your original legal difference of of your original legal system and then the the difference it system and then the the difference it system and then the the difference it had to make to ensure you know this was had to make to ensure you know this was had to make to ensure you know this was consistent with your morals.

consistent with your morals. consistent with your morals. And then this is a an example of And then this is a an example of And then this is a an example of overreach and in this case it found that overreach and in this case it found that overreach and in this case it found that it couldn't do the auto patch. It's it couldn't do the auto patch. It's it couldn't do the auto patch. It's talking about, you know, someone submits talking about, you know, someone submits talking about, you know, someone submits their DNA for uh genetic research, but their DNA for uh genetic research, but their DNA for uh genetic research, but the researcher finds they have a rare the researcher finds they have a rare the researcher finds they have a rare but treatable genetic disorder. Uh but but treatable genetic disorder. Uh but but treatable genetic disorder. Uh but currently your morals kind of say this currently your morals kind of say this currently your morals kind of say this shouldn't be allowed that they could shouldn't be allowed that they could shouldn't be allowed that they could disclose this disease to the the person disclose this disease to the the person disclose this disease to the the person submitting. And so this was raised to submitting. And so this was raised to submitting. And so this was raised to the user to me to kind of make a the user to me to kind of make a the user to me to kind of make a judgment. And then similarly, when you judgment. And then similarly, when you judgment. And then similarly, when you make the judgment, you get this this make the judgment, you get this this make the judgment, you get this this patch legal system. And so, you know, patch legal system. And so, you know, patch legal system. And so, you know, it's just sort of a a fun game and I it's just sort of a a fun game and I it's just sort of a a fun game and I posted on Twitter and shared it open posted on Twitter and shared it open posted on Twitter and shared it open source on GitHub. And for me at least, source on GitHub. And for me at least, source on GitHub. And for me at least, it was by far the most viral post I've it was by far the most viral post I've it was by far the most viral post I've had. And it sort of made me think like I had. And it sort of made me think like I had. And it sort of made me think like I think a lot of people just said it was think a lot of people just said it was think a lot of people just said it was sort of fun. You could stress test your sort of fun. You could stress test your sort of fun. You could stress test your morals, see if you have any interesting morals, see if you have any interesting morals, see if you have any interesting contradictions. But it also made me contradictions. But it also made me contradictions. But it also made me think, you know, is there maybe think, you know, is there maybe think, you know, is there maybe something more here? like could this be something more here? like could this be something more here? like could this be um you know have more like practical or um you know have more like practical or um you know have more like practical or bigger scope implications and so I'll bigger scope implications and so I'll bigger scope implications and so I'll just talk about three different branches just talk about three different branches just talk about three different branches I'm kind of exploring um the first and I'm kind of exploring um the first and I'm kind of exploring um the first and sort of leave leaving the legal area and sort of leave leaving the legal area and sort of leave leaving the legal area and really more practical is thinking about really more practical is thinking about really more practical is thinking about sort of an auto way to make sort of an auto way to make sort of an auto way to make constitutions for chat bots or really constitutions for chat bots or really constitutions for chat bots or really you know for agents in general where you you know for agents in general where you you know for agents in general where you know say you're a company and you want know say you're a company and you want know say you're a company and you want to have a um agent or chatbot that's to have a um agent or chatbot that's to have a um agent or chatbot that's customerf facing and you want it to sort customerf facing and you want it to sort customerf facing and you want it to sort of adhere to a moral code but also have

of adhere to a moral code but also have of adhere to a moral code but also have things it will and will not talk about. things it will and will not talk about. things it will and will not talk about. Um I've kind of in one branch formulated Um I've kind of in one branch formulated Um I've kind of in one branch formulated it so you in a similar way write your it so you in a similar way write your it so you in a similar way write your morals. You write what the chatbot morals. You write what the chatbot morals. You write what the chatbot should and not talk about and then it should and not talk about and then it should and not talk about and then it tries to write this codified system tries to write this codified system tries to write this codified system prompt and then you have these kind of prompt and then you have these kind of prompt and then you have these kind of adversarial agents trying to get it to adversarial agents trying to get it to adversarial agents trying to get it to either talk about something it shouldn't either talk about something it shouldn't either talk about something it shouldn't or refuse to talk about something it or refuse to talk about something it or refuse to talk about something it should. And I see it as this sort of should. And I see it as this sort of should. And I see it as this sort of analogy or analogous method to GEA um analogy or analogous method to GEA um analogy or analogous method to GEA um but really aimed at kind of building but really aimed at kind of building but really aimed at kind of building these codified system prompts. these codified system prompts. these codified system prompts. The uh second use case that I'm I'm The uh second use case that I'm I'm The uh second use case that I'm I'm particularly interested in is thinking particularly interested in is thinking particularly interested in is thinking of it as a way to sort of do more ad hoc of it as a way to sort of do more ad hoc of it as a way to sort of do more ad hoc or decentralized contracts. So I think or decentralized contracts. So I think or decentralized contracts. So I think in in a simple case say like you can in in a simple case say like you can in in a simple case say like you can specify how you want your data or specify how you want your data or specify how you want your data or privacy to be handled online and you can privacy to be handled online and you can privacy to be handled online and you can go through this sort of adversarial game go through this sort of adversarial game go through this sort of adversarial game to get this codified legal system of how to get this codified legal system of how to get this codified legal system of how you want your your data handled and if you want your your data handled and if you want your your data handled and if you go to you know say Apple releases a you go to you know say Apple releases a you go to you know say Apple releases a new terms of service or something you new terms of service or something you new terms of service or something you can run the um contradictions between can run the um contradictions between can run the um contradictions between your legal system and between Apple's your legal system and between Apple's your legal system and between Apple's terms of service and like surface any terms of service and like surface any terms of service and like surface any interesting contradictions or like interesting contradictions or like interesting contradictions or like synthetic cases where this would lead to synthetic cases where this would lead to synthetic cases where this would lead to a difference between how you know your a difference between how you know your a difference between how you know your morals, what you want and what the morals, what you want and what the morals, what you want and what the company is doing. And you know in the company is doing. And you know in the company is doing. And you know in the case that it's a big company, maybe you case that it's a big company, maybe you case that it's a big company, maybe you can't really change anything. It's not a can't really change anything. It's not a can't really change anything. It's not a negotiation, but you can at least be

negotiation, but you can at least be negotiation, but you can at least be sort of have better information about sort of have better information about sort of have better information about the contract you're signing. But I also the contract you're signing. But I also the contract you're signing. But I also think in the case of you know thinking think in the case of you know thinking think in the case of you know thinking more decentralized like if you're trying more decentralized like if you're trying more decentralized like if you're trying to have contracts without you know some to have contracts without you know some to have contracts without you know some central authority kind of enforcing them central authority kind of enforcing them central authority kind of enforcing them and you're trying to maybe do contracts and you're trying to maybe do contracts and you're trying to maybe do contracts across different countries. Um, thinking across different countries. Um, thinking across different countries. Um, thinking about like if you can specify your about like if you can specify your about like if you can specify your morals and how you want to like morals and how you want to like morals and how you want to like interface, you know, maybe it's just interface, you know, maybe it's just interface, you know, maybe it's just like contracted work, how you want your like contracted work, how you want your like contracted work, how you want your work to be paid for and and and the work to be paid for and and and the work to be paid for and and and the different morals surrounding that. And different morals surrounding that. And different morals surrounding that. And the other party can do the same. And the other party can do the same. And the other party can do the same. And then you both get this kind of stress then you both get this kind of stress then you both get this kind of stress tested codified contract. And then you tested codified contract. And then you tested codified contract. And then you can kind of find the the disagreements can kind of find the the disagreements can kind of find the the disagreements if there are any and surface them before if there are any and surface them before if there are any and surface them before you agree. And then you can kind of be you agree. And then you can kind of be you agree. And then you can kind of be more confident in the the contract as a more confident in the the contract as a more confident in the the contract as a whole. whole. whole. And the last thing and maybe the kind of And the last thing and maybe the kind of And the last thing and maybe the kind of more aspirational angle is thinking more aspirational angle is thinking more aspirational angle is thinking about smarter government or more about smarter government or more about smarter government or more efficient government. Um I think there efficient government. Um I think there efficient government. Um I think there would be a lot of different privacy would be a lot of different privacy would be a lot of different privacy issues and logistical issues but sort of issues and logistical issues but sort of issues and logistical issues but sort of ignoring those for now and just thinking ignoring those for now and just thinking ignoring those for now and just thinking big picture. I think for voters or big picture. I think for voters or big picture. I think for voters or constituents, you know, this could be a constituents, you know, this could be a constituents, you know, this could be a really interesting way if you you really interesting way if you you really interesting way if you you defined your morals, you have this defined your morals, you have this defined your morals, you have this stress- tested legal code, sort of any stress- tested legal code, sort of any stress- tested legal code, sort of any new bill or politician that comes out, new bill or politician that comes out, new bill or politician that comes out, you could kind of run your contract you could kind of run your contract you could kind of run your contract against theirs and surface, you know, against theirs and surface, you know, against theirs and surface, you know, what are the the cases you would

what are the the cases you would what are the the cases you would disagree or or interesting disagree or or interesting disagree or or interesting uh points that are kind of immoral to uh points that are kind of immoral to uh points that are kind of immoral to you or or a contradiction. Um, I also you or or a contradiction. Um, I also you or or a contradiction. Um, I also think, you know, relating to one think, you know, relating to one think, you know, relating to one another, it's like a more I think we all another, it's like a more I think we all another, it's like a more I think we all have a lot of nuance in the way we feel have a lot of nuance in the way we feel have a lot of nuance in the way we feel about things and this is a way to kind about things and this is a way to kind about things and this is a way to kind of get to that nuance instead of arguing of get to that nuance instead of arguing of get to that nuance instead of arguing over just values which is, you know, over just values which is, you know, over just values which is, you know, often the values are not in often the values are not in often the values are not in contradiction. And then I think maybe a contradiction. And then I think maybe a contradiction. And then I think maybe a little more practically for legislators, little more practically for legislators, little more practically for legislators, you could imagine um if you want to you could imagine um if you want to you could imagine um if you want to propose a bill and you can have like a a propose a bill and you can have like a a propose a bill and you can have like a a simulation of all the other legislators simulation of all the other legislators simulation of all the other legislators and a and a legislative body, you could and a and a legislative body, you could and a and a legislative body, you could sort of stress test it before sort of stress test it before sort of stress test it before submission. And so the the third branch submission. And so the the third branch submission. And so the the third branch I've been building on this project is I I've been building on this project is I I've been building on this project is I I tried to do this for the US Senate. I tried to do this for the US Senate. I tried to do this for the US Senate. And so what I did is I first had Claude And so what I did is I first had Claude And so what I did is I first had Claude go through all current US senators and go through all current US senators and go through all current US senators and look at, you know, kind of all their look at, you know, kind of all their look at, you know, kind of all their voting history and anything else that voting history and anything else that voting history and anything else that was public and build their kind of moral was public and build their kind of moral was public and build their kind of moral system and then ran it through the system and then ran it through the system and then ran it through the loophole process to get a codified sort loophole process to get a codified sort loophole process to get a codified sort of legal code.

of legal code. of legal code. And then on this system, you can, you And then on this system, you can, you And then on this system, you can, you know, take any current bill that's being know, take any current bill that's being know, take any current bill that's being proposed or even propose your own and proposed or even propose your own and proposed or even propose your own and submit it. And you can have Claude sort submit it. And you can have Claude sort submit it. And you can have Claude sort of simulate how each senator would vote. of simulate how each senator would vote. of simulate how each senator would vote. And so here you can see like a breakdown And so here you can see like a breakdown And so here you can see like a breakdown of um some senators, which way they're of um some senators, which way they're of um some senators, which way they're leaning and sort of their reasoning leaning and sort of their reasoning leaning and sort of their reasoning behind the vote. behind the vote. behind the vote. Um and I think you know it's sort of Um and I think you know it's sort of Um and I think you know it's sort of interesting just to think about like interesting just to think about like interesting just to think about like seeing what you know a proposed piece of seeing what you know a proposed piece of seeing what you know a proposed piece of legislation how people would vote but legislation how people would vote but legislation how people would vote but also this sort of becomes and I think on also this sort of becomes and I think on also this sort of becomes and I think on theme of the conference its own theme of the conference its own theme of the conference its own verifiable domain or loop and you could verifiable domain or loop and you could verifiable domain or loop and you could think about even kind of hill climbing think about even kind of hill climbing think about even kind of hill climbing the bill towards um getting like a super the bill towards um getting like a super the bill towards um getting like a super majority or whatever you need it to majority or whatever you need it to majority or whatever you need it to pass. And so in this case, like this pass. And so in this case, like this pass. And so in this case, like this this Medicare bill I was testing, you this Medicare bill I was testing, you this Medicare bill I was testing, you know, it found that I think it know, it found that I think it know, it found that I think it originally started at like a 5050 vote originally started at like a 5050 vote originally started at like a 5050 vote and it found ways to hill climb the and it found ways to hill climb the and it found ways to hill climb the language of the bill such that it passed language of the bill such that it passed language of the bill such that it passed with um 52 votes. And I think, you know, with um 52 votes. And I think, you know, with um 52 votes. And I think, you know, this is an example of it can find um this is an example of it can find um this is an example of it can find um like the sort of the core tenants of the like the sort of the core tenants of the like the sort of the core tenants of the bill and it can try to find like run the bill and it can try to find like run the bill and it can try to find like run the the bill against each senator's contract the bill against each senator's contract the bill against each senator's contract and find is there any way I can change and find is there any way I can change and find is there any way I can change the language such that I don't violate the language such that I don't violate the language such that I don't violate sort of the core tenants or morals of sort of the core tenants or morals of sort of the core tenants or morals of the bill and kind of do those auto the bill and kind of do those auto the bill and kind of do those auto patching that way. And then it can also patching that way. And then it can also patching that way. And then it can also find um you know kind of rank order the find um you know kind of rank order the find um you know kind of rank order the changes that would need to be in place

changes that would need to be in place changes that would need to be in place to maximize votes and you as a user can to maximize votes and you as a user can to maximize votes and you as a user can kind of choose the trade-offs that way. kind of choose the trade-offs that way. kind of choose the trade-offs that way. And then the last thing I I've been And then the last thing I I've been And then the last thing I I've been trying out more recently with this trying out more recently with this trying out more recently with this branch is actually looking at, you know, branch is actually looking at, you know, branch is actually looking at, you know, kind of even bigger picture, like can kind of even bigger picture, like can kind of even bigger picture, like can this lead to an even more efficient of this lead to an even more efficient of this lead to an even more efficient of government where you have every sort of government where you have every sort of government where you have every sort of constituent in a in a state or whatever constituent in a in a state or whatever constituent in a in a state or whatever the district is sort of have their legal the district is sort of have their legal the district is sort of have their legal code and then you could just submit any code and then you could just submit any code and then you could just submit any bill and actually measure sort of the bill and actually measure sort of the bill and actually measure sort of the agreement between like the the actual agreement between like the the actual agreement between like the the actual voters. And so for this, I took the voters. And so for this, I took the voters. And so for this, I took the Nvidia has this really great data set of Nvidia has this really great data set of Nvidia has this really great data set of USA personas. And so I took 500 personas USA personas. And so I took 500 personas USA personas. And so I took 500 personas per state and it's supposed to be sort per state and it's supposed to be sort per state and it's supposed to be sort of well representative of the state's of well representative of the state's of well representative of the state's population. Did the same process of population. Did the same process of population. Did the same process of having them given the persona, draft having them given the persona, draft having them given the persona, draft their morals, draft their sort of legal their morals, draft their sort of legal their morals, draft their sort of legal contract, and then take any bill you're contract, and then take any bill you're contract, and then take any bill you're interested in and kind of run it against interested in and kind of run it against interested in and kind of run it against each state. And you can also, you know, each state. And you can also, you know, each state. And you can also, you know, measure how much people like this bill measure how much people like this bill measure how much people like this bill or how much it's in agreement with their or how much it's in agreement with their or how much it's in agreement with their morals and then also do this hill morals and then also do this hill morals and then also do this hill climbing where you kind of optimize the climbing where you kind of optimize the climbing where you kind of optimize the bill for the people.

bill for the people. bill for the people. And so just to conclude, you know, at at And so just to conclude, you know, at at And so just to conclude, you know, at at minimum, I think it's a pretty fun game. minimum, I think it's a pretty fun game. minimum, I think it's a pretty fun game. I'm biased, but it's a lot of fun to I'm biased, but it's a lot of fun to I'm biased, but it's a lot of fun to just try out different um you know, just try out different um you know, just try out different um you know, things you care about, put in your things you care about, put in your things you care about, put in your morals, see if there's any morals, see if there's any morals, see if there's any contradictions. you know, often it will contradictions. you know, often it will contradictions. you know, often it will raise some really interesting questions raise some really interesting questions raise some really interesting questions and then once you kind of provide that and then once you kind of provide that and then once you kind of provide that nuance, the game, you know, the the nuance, the game, you know, the the nuance, the game, you know, the the agents won't be able to find any more agents won't be able to find any more agents won't be able to find any more contradictions and you can kind of feel contradictions and you can kind of feel contradictions and you can kind of feel good that you have like a a consistent good that you have like a a consistent good that you have like a a consistent nuanced uh moral system. Um, but I I am nuanced uh moral system. Um, but I I am nuanced uh moral system. Um, but I I am interested in, you know, exploring could interested in, you know, exploring could interested in, you know, exploring could this be are there kind of real this be are there kind of real this be are there kind of real applications here for some kind of like applications here for some kind of like applications here for some kind of like decentralized or better contracts and decentralized or better contracts and decentralized or better contracts and maybe even for legislators as a way to maybe even for legislators as a way to maybe even for legislators as a way to sort of stress test your bills and even sort of stress test your bills and even sort of stress test your bills and even think about how to write um better laws think about how to write um better laws think about how to write um better laws that are, you know, better for the that are, you know, better for the that are, you know, better for the people in your district or more people in your district or more people in your district or more representative of what the people in representative of what the people in representative of what the people in your district want. Um, and so this QR your district want. Um, and so this QR your district want. Um, and so this QR code is to the the Senate simulator. So code is to the the Senate simulator. So code is to the the Senate simulator. So I encourage you if you're interested to I encourage you if you're interested to I encourage you if you're interested to play and um the other one is to my play and um the other one is to my play and um the other one is to my website which has the the full GitHub to website which has the the full GitHub to website which has the the full GitHub to loophole um and please you know play loophole um and please you know play loophole um and please you know play with it fork it uh I'd love to have with it fork it uh I'd love to have with it fork it uh I'd love to have other contributors. Thank you.