Glen Weyl, RadicalxChange Foundation founder and political economist & social technologist at Microsoft Special Projects, and Puja Ohlhaver, strategist at Flashbots, discuss “soulbound tokens” and their implications for collaboration and social organization in a variety of spaces. Topics covered include:
- how Glen got involved in crypto, wrote a book, and came to co-write a paper with Vitalik Buterin
- how Puja studied economics and got in touch with Glen in pursuit of a middle ground between left and right politics
- what soulbound NFTs are and how they work
- how Vitalik’s paper articulated how the concept of the soulbound token could advance decentralized collaboration in web3
- how decentralized reputation can enable larger networks of coordination
- how identity can be understood through the lens of community participation
- how the name ‘soulbound token’ came about
- how these tokens could provide a technology for those who value scarcity but disdain speculation
- how identity-dedicated tokens could work on a technical level (including recovery)
- what is DeSoc and why is it important?
- how non-transferable tokens can improve DAO organization, including resolving issues with quadratic funding
- how identity-locked tokens can be protected from bots and AI abuse
- how these tokens could support community privacy and responsible information disclosure
Thank you to our sponsors!
Crypto.com: https://crypto.onelink.me/J9Lg/unconfirmedcardearnfeb2021
Beefy Finance: https://beefy.finance/
Related Reading
- Decentralized Society: Finding Web3’s Soul (Weyl, Ohlhaver, and Buterin; 2022)
- Georg Simmel, German neo-Kantian who laid groundwork for antipositivism
- How Blockchains Can Help Create Little Democracies Everywhere
Soulbound Tokens
- Vitalik on Soulbound tokens
- TL;DR written by Glen
- Other write-ups
Projects Mentioned
Glen Weyl
Puja Ohlhaver
Episode Transcript
Laura Shin:
Hi, everyone. Welcome to Unchained. Your no-hype resource for all things crypto. I’m your host Laura Shin, author of the Cryptopians. I started covering crypto seven years ago and as a senior editor at Forbes was the first mainstream media reporter to cover cryptocurrency full-time. This is the June 7, 2022 episode of Unchained.
With the Crypto.com app you can buy, earn, and spend crypto in one place. Download and get 25 dollars with the code Laura, link in the description.
This episode of Unchained is brought to you by Beefy Finance, the multi-chain yield optimizer. Beefy is the easiest way to earn more from your crypto. Deposit funds into Beefy’s secure vault to auto-compound yields across 12 blockchains. Got crypto? Choose Beefy.
Today’s topic is Soulbound Tokens. Here to discuss are Glen Weyl, founder of the Radical Exchange Foundation, and political economist and social technologist at Microsoft Research Special Projects, and Puja Ohlhaver, strategist at Flashbots. Welcome, Glen and Puja.
Puja Ohlhaver:
Thanks for having us.
Glen Weyl:
Happy to be with you.
Laura Shin:
Quick note that I am in Europe and do not have the cord that connects my nice mic to my computer, so my sound is not great. Apologies. And I also need to note that because I have COVID or at least I had it a few days ago. I don’t know about now, I may also cough during the episode. I likely will. Hopefully, I will not choke on screen. We’ll see.
All right. First, let’s discuss each of your backgrounds and how it is that you came to write this paper with Ethereum creator, Vitalik Buterin, on Soulbound Tokens. Glen, why don’t we start with you?
Glen Weyl:
So, I have a background as an economist. I was a professor at the University of Chicago for a few years and then I ended up at Microsoft mostly for personal reasons. I was developing a lot of sort of radical ideas about how to reimagine our politics and economy. In 2016, when a lot of the populist wave started to happen in mainstream politics I decided to write a book called Radical Markets that put those things together, jointly with Eric Posner, into sort of a vision of addressing some of the big picture problems in the world.
I was not at all involved in what was then called the blockchain space. In fact, I was quite skeptical of what little I knew about it. But the book ended up not being a particularly great mainstream success, but found an audience within the blockchain world and actually one part of that all happening was the great episode that we all did with Santi Siri a few years back.
Laura Shin:
Yeah. That was a great episode. I will link to it in the show notes for people.
Glen Weyl:
Yeah. And Vitalik and I got to be pretty close through that process. We’ve written several things together, and we founded The Radical Exchange Foundation together. He’s been the primary financial supporter of The Radical Exchange Foundation, and he’s on the board. We’ve just been in conversation about ways to integrate technology with reimagining new political and economic structures since then. In January, he wrote a piece about Soulbound Tokens. As we might get into, there wasn’t anything that was a huge technological breakthrough there, and we can talk about that, but it made me see ways that there was a much quicker path from the current state of the Web3 ecosystem towards some of the things we’ve been imagining.
Puja and I started thinking about that and we wrote a paper around that, and we ran it by Vitalik and he was enthusiastic and contributed as well. That’s how this paper came about.
Puja Ohlhaver:
Well, I’m actually a lawyer by training and my first sort of career was in public policy, thinking about issues. This was around 2003-2007 and then I went to law school during the financial crisis and studied financial markets pretty deeply and then became an entrepreneur, stumbled across Glen’s book, DMed him on Twitter, thought this was really interesting. I thought a lot of the ideas he proposed was kind of a third way between the left and the right and resolved a lot of interesting issues while raising new issues. Through Glen, I became introduced to Radical Exchange and Audrey Tang, who I found to be a total inspiration in her movement in digital democracy in Taiwan.
So, that has been my entry point more from the social, political, innovation lens and also in crypto as well. I’ve always been interested in the social and political ramifications of crypto.
Laura Shin:
So, let’s just give a really basic definition of today’s topic, Soulbound Tokens and also talk about where this concept originates from.
Glen Weyl:
Yeah. So, Soulbound Tokens was an idea that Vitalik discussed in January and he basically described it as just a non-transferrable token, a non-transferrable NFT basically. This has a bunch of different types of antecedents, one of which is World of Warcraft and the fact that there’s lots of items in World of Warcraft that are not just tradeable swords or something like that but that are actually attached to a character, like wings. That was Vitalik’s inspiration.
There’s an enormous community of people who’ve been working on things related to decentralized identity for years who have been thinking about things that have that basic structure for a long time. We can get into some of the details of that later, but this attempt to connect it to all the exciting energy around new forms of social and economic organization within the Web3 community was really what I think the contribution that Vitalik’s paper made was, and that’s really what we tried to emphasize in this paper.
So, Soulbound Tokens is more of a general allusion to this notion of these non-transferable statements about people that are at least partly public so they can interact with the apps. It’s not really a full technical specification, and the emphasis of the paper, at least as we intended it, was to be on opening people’s eyes to both the near-term and really ambitious longer-term applications that these could enable and sort of painting a sociotechnical picture there rather than to focus primarily on the design of a particular standard. Which, as I mentioned, there’s been a lot of work on for some time and there’s continuing suggestions about it. Happy to discuss that but the intention of the paper is really around illustrating all the things that this can make possible when it interacts with the energy in the Web3 ecosystem.
Laura Shin:
Puja, do you want to flush out what that vision is?
Puja Ohlhaver:
Yeah. So, it’s a vision, first of all, of bottom-up coordination starting first with individuals nested within their communities and forming broader, nested forms of cooperation. So, there’s a few principles around our ideas. One basic one is just local control and the idea that those who are closest to a problem generally have a larger stake in its resolution. So, what we try to do is paint a picture of how, starting with decentralized reputation and communities these communities can kind of compose into larger and larger networks of coordination in a consensual, yet decentralized and pluralistic way.
Yeah. I mean we go into a lot. We go to the far reaches of prediction markets and AI, but ultimately, the idea is to ground these technologies in the people that use them and have networks coordinated and owned by the people that use them and the communities that participate in them in compulsive ways. Maybe, Glen, do you want to take a shot at it?
Glen Weyl:
One other sort of philosophical element I would add is there’s a sociologist named Georg Simmel who was one of the founders of the field but got kind of forgotten in the ’50s. He had this idea that there is no such thing as an individual in primitive or early human societies. That people are part of a group and that individualization happens as a result of modernity because people become parts of many different social communities. Your work life is different from your home life, is different from your religious life, is different from etcetera, etcetera. As you accumulate all those social relationships, sort of the pattern of social relationships becomes unique for each person. This was the inspiration for what became social graph theory and the idea of social networks.
I think that underlying vision, the notion that that network or relationships, that the individual actually comes out of the collection of communities that she’s a part of and that communities emerge from the set of individuals within them but that that underlying network of connections is the fundamental thing. It was the inspiration for the internet. It was the inspiration for so much of how we live and yet it’s never really been expressed in the way that our technologies are setup, and we’re trying to describe a path from the current Web3 ecosystem to something like that that you might think of as sort of like the internet society that was imagined even before the technologies for it were possible.
Puja Ohlhaver:
Yeah. I just want to add on that quickly. So, in the paper, we talk a lot about how individuals are differentiated by their communities and vice versa, communities are differentiated by their individuals. So, there are two ways to look at this. You can look at it as some people sort of got caught up in the framing around Soulbound Tokens. You can also call them community-bound tokens because it’s the communities in which you participate in that really define you and that’s what Glen was just alluding to.
Laura Shin:
Yeah. Actually, let’s talk about the name because a lot of people…I tweeted on Twitter what I should ask you and multiple people asked about the name, and I agree. Even just reading through the paper where it says your account will be called your Soul. It almost feels like it was purposely written to mimic some futuristic dystopian science fiction. I was a little bit like, why not just call it an account? So, why is it called Soulbound Tokens and why are you suggesting the account would be called your Soul?
Puja Ohlhaver:
It’s kind of interesting. I think the folks who have kind of come from a more pluralistic background and see themselves with a nested cooperation network, Soulbound really resonated with them. I think folks who have a sort of more individualistic, atomistic perspective, Community-bound resonates more with them. But I think frankly, it was just a lot of fun to write the paper and riff off of the metaphor. That was part of it. I mean as I was writing it, I was having a lot of fun.
But the other idea is, I get the soul can sound dystopic, but souls are actually things you’re supposed to be very careful about. You know, you don’t want to just sell your soul. You don’t want to financialize your soul, and so the idea was more around, okay, these are very careful…your soul is something very careful and precious and you want to shepherd it carefully and not financialize it into some sort of dystopic reality. But Glen, what do you think?
Glen Weyl:
I mean I think part of the intention, frankly, was to do something whimsical and deliberately not something that sounds like a technical standard so that people wouldn’t think that we were proposing a technical standard that should be called Soulbound or Souls or whatever. We were trying to be more metaphorical and more elusive, deliberately. Like now there’s pluses and minuses of that. Obviously, there’s a lot of people who’ve taken it very, very literally and that’s unfortunate. I regret that at least. But the intention, at least from my part was precisely to illustrate and allude to possibilities rather than fully specify technical standards, which if you use a much more technical term would have sounded that way.
Laura Shin:
Okay. Yeah. It’s funny because I was definitely going to ask about this, but I wasn’t going to have it so high up in the conversation. But between the fact that it just naturally came up and then, like I said, noticing on my tweet how many times that particular point got mentioned I was like, you know what, this is something people want to know about. Another thing I think that is kind of hard to see at this moment when it’s this concept that’s been thrown out there or there’s been a lot of examples given. We’re already seeing there’s at least one token project that is looking to use this concept. I’m sure it’s just at this phase right now where a lot of people are like what kinds of problems exist that this is going to solve. So, can you talk a little bit about what problems it is that you see that you feel that this is a solution that makes sense for those?
Glen Weyl:
Let me just start with things that are very, very near-term and concrete. These are not the things that excite me the most but I think it’s important that we start with them because otherwise it sounds very abstract and futuristic. So first of all, there are a lot of organizations that would like to issue NFTs for more or less the same purposes that everyone’s issuing NFTs. Basically, to allow for some meaningful provenance on some object that is in some sense digitally unique or scarce or something like that but who think speculation is awful and don’t want to be throwing stuff out there that’s going to be the subject of speculative success.
They want them to be purchased by people who are actually going to value them for what they are rather than just seek to resell them. I have another design that does a similar thing to that while still allowing for transferability, but I think that a non-transferable NFT is a very simple and natural way to just address that concern, and I think it might unlock many more mainstream organizations getting all the benefits of NFTs.
Just to take one small step beyond that, if you’re an artist who’s issuing an NFT, right now you commit to that being scarce by issuing a press release, usually. You say this is going to be part of a limited edition or whatever it is that you want to say about it. An alternative would be that you could issue a Soulbound Token to your wallet, Soul, whatever you want to call it that is sort of countersigning or issued by the transferrable NFTs that you’re putting out there. That is a public commitment to effectively not double spend in the sense that you’re not going to recreate more than however many you said you would of that object.
So, I mean these are very simple things, right. There’s nothing amazing or earthshattering about any of this, but I do think that there are things that are really missing from the ecosystem that could easily be solved using this type of an approach.
Laura Shin:
So some examples would be like your driver’s license, your university degree.
Glen Weyl:
No. Let’s not go there. Look, I’m not saying that those things couldn’t be included here as well, but I certainly wouldn’t want driver’s license to be on chain, and we really have to figure out the privacy model before you want that. University degree, you know, a lot of people put that on LinkedIn anyway, right.
Laura Shin:
A lot of people lie about it on LinkedIn.
Glen Weyl:
Yeah. It would be nice, I think, if people could have more transparency about what’s actually true there. This would be helpful in that regard.
Laura Shin:
But why did you not want to talk about actual examples? Because I feel like this conversation is a little bit difficult to have when it’s so abstract, and I feel like it’s actually easier to understand through concrete examples.
Glen Weyl:
I don’t want to talk about driver’s licenses because I don’t want those to be in a model where our primary idea is that this is public. So, university degrees I’m reasonably comfortable with. I’m more comfortable with things like I just said before, which are NFTs that are not transferrable or social tokens.
Laura Shin:
But why would I need that? What would I use it for?
Puja Ohlhaver:
I’ll give you an example, Laura. Initially, when we wrote the paper we assumed, say, take just a small set of tokens which people or memberships and affiliations, people would be willing to have public about themselves anyway. And so where is there a degree of high publicity and also a problem of, say, monopolistic power? Science is actually a really great starting point for that. Scientists are very public about their credentials. They’re very public about their journals and publications, and yet scientists are like a really exploited group of sort of the academic class in the knowledge productions supply chain. They submit their articles to journals, they don’t get paid, the journals are major rent extractors. Same thing with peer reviewers.
So, you can imagine a small scientific community getting together and representing their credentials, which they do publicly anyway, and their publication’s public anyway, and forking away, say, like a rent extractive, journalistic monopoly like Nature or Cell or something like that and recreating that and forking away those monopoly rents and sort of regenerating them, redistributing them back into the scientific networks to produce more scientific knowledge and having scientists paid and peer reviewers paid. So, that’s like a very low-hanging fruit application I see.
Another really important application to the crypto space, in general, is getting a sense of actually how decentralized some of our systems are and it’s really important that we do that because the whole sort of premise and justification for crypto is decentralization. It’s what protects it from being regulated by antitrust and also securities regulation as well. And so, part of this paper was to help us give us some primitives to start to measure up the decentralization of our systems and justify them. I think we can actually apply the same analysis to traditional finance and have some really interesting conversations about the ways in which Web3 might be superior or even leading traditional financial networks that tend to also concentrate power.
So, I think there are a lot of applications, but I would say I’m very bullish on the scientific network and research case.
Glen Weyl:
Another example that’s very similar to that that I would give is open source software contributors. Again, extremely public. Like, the whole point is for it to be public, open, social capital in a public way. Not saying everything in open source is that way but that’s the way that the dominant ethos of the community is. Soulbound Token type constructs would be very natural for roles that you have within an open source repo like Maintainer, Contributor, User, etcetera. Those are already public, effectively, badges that usually GitHub gives out and a decentralized approach to them would allow a bunch of really interesting additional applications.
Moreover, even though I was not concerned about driver’s licenses, I was concerned about that specific example you gave. It’s not that there aren’t a ton of examples that are credential-like that I would be comfortable with. No one puts their driver’s license on their LinkedIn profile. Almost no one puts their driver’s license on their Twitter profile. When we talk about things that are public and we’re starting from a public perspective, and we can talk about privacy. That’s something we’re really interested in. Starting from a public perspective, anything that you don’t want to put on LinkedIn or Twitter I would not say belongs in a public setting. There’s lots of things that you do put in those contexts, and they are pretty centralized right now and I think we’d get a lot of value out of decentralizing those types of things. Choose any example that fits those cases and I’m happy to talk about them.
Laura Shin:
On a basic technical level how does a Soulbound Token work? Because obviously, you would have to have some mechanism by which you keep people from either transferring the token or you have to account for situations in which maybe they’re switching between wallets. How do they keep the Soulbound Token or what if something happens where one of their wallets gets compromised and one of the Soulbound Tokens was in there. So just in general, how do you kind of account for these different scenarios?
Puja Ohlhaver:
So, a key to Soulbound Tokens, you actually can’t have Soulbound Tokens unless you also have this mechanism we describe called community recovery. The idea around community recovery is that you basically are the intersexual vote of your social network. So, if you try to sell your wallet with your non-transferrable tokens, you couldn’t actually credibly do that because you could always just recover it back from your community. So community recovery as a wallet mechanism is very key to Soulbound Tokens. The two really go hand in hand and we talk a little bit about it at the end of the paper. How do you bootstrap both of these things and it’s a really important area, probably the most important area of research and it’s the hardest area of research.
So, when folks DM me and they’re super excited about Soulbound Tokens I’m like okay, okay, that’s great. I’m interested in community recovery right now because the concept really needs that to take off.
Glen Weyl:
Yeah. The only slight way to disqualify what Puja just said is that we don’t have community recovery yet, and these bootstrapping things are going to have to help us get there. And so there’s going to be things that are going to be sort of proto Soulbound Tokens that will exist prior to community recovery work. I’m open to a variety of those and we discuss a bunch of different paths in the paper. But at some level, as Puja pointed out, those aren’t real Soulbound Tokens until you have a community recovery mechanisms that anchor them.
Laura Shin:
Yeah. So, what are some ideas for structuring the community recovery? This was not the first time I had heard somebody talk about these ideas. So, I feel like they’re sort of percolating out there now, so let’s say that I am the person with the Soulbound Tokens. I either lost the keys to my wallet or somebody phished them from me or whatever it is. How would this work?
Puja Ohlhaver:
So, a bunch of different ways to think about it. We talk about correlation discounts in the paper, so maybe you want to apply some correlation discounts. Like the mechanism would apply correlation discounts to…
Glen Weyl:
Puja, before we even get there, I think we should describe the basics of what it is at all. So, is that right, Laura?
Laura Shin:
Yeah.
Glen Weyl:
The basic idea, Laura, is let’s start with the idea of social recovery and let’s explain that and then let’s move to community recovery. So, the idea of social recovery is that you appoint a set of guardians, other wallets, and some sort of qualified majority of them, let’s say three out of five or whatever, can always come together and unlock your keys and allow you to regenerate your private key. And perhaps, even you put some safeguards in there like if you’re going to make a really large transaction or if you’re going to make a bunch of transactions in rapid succession or something like that, you need support from some qualified majority of those guardians, let’s say.
So, that is already pretty well established. I think it’s generally working fairly well, at least relative to other non-custodial approaches within the ecosystem right now. The origin and loop ring wallets both have it. So, that’s great. That’s a great mechanism and we think it’s something to build on.
Laura Shin:
Can you give examples of who the different entities would be or what types of entities they would be?
Glen Weyl:
Well, usually in the Web3 ecosystem, so far it’s other friends. It’s just individuals who you have a relationship with and whose address you know. But community recovery is a somewhat different paradigm than that. I mean of course, it could be DAO or something. It can be any wallet, so I don’t know the distribution right now of when people use these social recovery mechanisms, how often it’s a DAO or group or a foundation versus how often it’s an individual. My understanding is it’s usually a set of individuals who you have a relationship with.
What we’re hoping to do to go beyond that is to deal with what I see as some of the limitations of the social recovery mechanism. Number one, you need to make sure all the people you’re linking to, like, themselves, haven’t been compromised, that they’re still your friends. That as your relationships evolve that you evolve that set of…so, it’s somewhat intensive for the user in a way that its relevance only becomes clear when you have a compromising event. I think there’s a significant concern that that isn’t really going to do a great job of…again, it’s much better than other purchased non-custody and I’m a big fan of it, but I don’t think it’s going to necessarily do the best job keeping up with the evolving nature of people’s lives.
What we propose instead, and it’s somewhat vague, but I think you can get a sense for it, is that the Soulbound Tokens that you have, which denote affiliations, relationships, public elements of employment, etcetera, would be a substrate for choosing a set of people who are your guardians in the social recovery in an automated way. So, let me turn to Puja’s example of the scientific, and just be super, super concrete. This is very, very specific but imagine that you have a Soulbound Token for every paper that you’ve coauthored and there is a fungible but not transferrable equivalent Soulbound Token that’s held by every other author of each of those papers. When you want to recover your wallet, the system randomly chooses five papers that you’ve authored and asks you to get at least one coauthor of each of those papers or maybe three of those papers, whatever. There’s some bunch of floating parameters here but some parameters that you have to get those people to come and recover your wallet.
The nice thing about that is that stays up to date as you publish new things, as your interests evolve. It’s done in an automated way that doesn’t require you to constantly maintain those things, and it can be enriched in a bunch of ways that Puja was getting at. Puja, I don’t know if you want to turn to the correlation discount point.
Puja Ohlhaver:
Right. So if you imagine Soulbound Tokens representing your memberships and affiliations, if you want wallet recovery to be secure, you don’t want, for example, recovery to happen with a set of collusive individuals who could, say, highjack your wallet, right. You want it to be amongst the maximally uncorrelated diverse members from different communities who aren’t talking to each other, right, and who can’t collude to steal your wallet from you, right.
And so, SBTs kind of give you a natural substrate to actually calculate sort of correlation discounts or maximally uncorrelated individuals in real time to have wallet recovery. Then there’s a lot of…it’s a lot of research. I mean it’s nothing easy. I’m not an expert at this but there’s recency bias and other kinds of things you need to take into account, and it will take I don’t know how long, but it’s not going to be solved in the next three months or something, right?
Glen Weyl:
I think that the ideal version of this, Laura, is one where it automatically generates a set of guardians for you effectively whose only point of contact is through you. You’re sort of the network central node in the set of guardians that are created by the generation. I’m not saying this will work for everyone, but I can give very clear illustrations from my life. So, I used to be an academic, as I mentioned. I was in the field of economics. I have all kinds of friends from the world of economics. I would never trust those people to be the guardians of something for me because they represent one element of my life, and I’m not that tied to the economics community anymore, and I would be worried about a potential of those people to screw with me in some way.
I also have a bunch of crypto people that I hang out with. There’s essentially zero overlap between the crypto people and the econ people, and again, I might be worried that those people could get together. There’s all kinds of weird stuff that goes on in the crypto world, you know what I mean, and I wouldn’t want those people alone to do it. I also have a bunch of collaborators from philosophy and from ecology because I’ve written in those fields, but those people don’t talk to each other at all. Their point of contact is through me, you know. So, if you had a system that autogenerated a set of guardians that was drawn from those maximally noncommunicating fields that maximally have the property that I’m the central node between them, that seems about as secure of a way as you could come up with to do my social recovery scheme and not something that you’d want individuals to have to do the calculation of themselves. That’s one of the problems with social recovery is that its sort of relying on them to do that.
Laura Shin:
Yeah. I can actually already picture which groups I would use for my recovery because it just feels like these are the natural groups where I’m close to all of them but they don’t know each other. I’m sure most people could do that exercise, so it’s not impossible to see how this might someday be applied. So, in a moment, we’re going to talk about things like decentralized society and kind of different ways that these might be used but first a quick word from the sponsors who make the show possible.
Finance is changing, strategies are changing, holding is changing. Beefy Finance, the multichain yield optimizer, allows you to maximize passive income while you sleep. Simply deposit your crypto into Beefy’s secure, industry-leading, auto-compounding vaults to put your funds to work. Each one of Beefy’s 740 vaults automatically reinvests the interest gained on your crypto deposits earning you more while saving you time and fees. Beefy’s strategies create bank busing APYs with zero percent deposit fees at the click of a button. Join 1.4 billion dollars of investments and understand why so many users trust Beefy with their financial independence. Visit Beefy.finance and take control of your financial future.
Join over 10 million people using Crypto.com, the easiest place to buy, earn, and spend over 150 cryptocurrencies. Spend your crypto anywhere using the Crypto.com Visa card. Get up to 8 percent cash back instantly, plus 100 percent rebates for your Netflix, Spotify, and Amazon Prime Subscriptions. Download the Crypto.com app now and get 25 dollars with the code Laura, link in the description.
Back to my conversation with Glen and Puja. So, as you wrote about in your white paper, Soulbound Tokens are just a piece of this larger vision you have that you call decentralized society or DeSoc. What is that and why is that important?
Puja Ohlhaver:
So, as I mentioned earlier, there’s a few principles of DeSoc. I would say two key ones. One is local control and that’s the idea that local communities who have the greatest stake in our resolution of a problem should have control over solving that problem but also this other principle called cooperation or consensus across difference. We write a lot about that, and we provide some mathematical representations of that, but the idea basically is that when you want to compose communities into broader cooperating networks to provide broader shared goods, you want to draw consensus from the maximally diverse participants, and that’s a better signal for shared goods than, say, consensus among a group of individuals who might be correlated in some way and might be over-coordinated more may be colluding in some way.
And so, decentralized society is really a picture of how we can get these communities to pluralistically cooperate into very rich social networks of broader coordination and generate broader and broader goods, a shared goods at different levels in a very dynamic, pluralistic way.
Laura Shin:
There are a few kind of different applications that you say that you feel Soulbound Tokens could open up in the crypto world. There’s one example of a financial one, which is under collateralized lending. So, how would this work with Soulbound NFTs?
Glen Weyl:
So, the basic idea is that if people want to stake some reputation that they have on something, and this can be to get a financial loan or it could be in the case of the artists to guarantee that they won’t issue something else very similar to this in the future. Anytime you want to stake non-financial reputation you need some representation of what that reputation is and you need to make some sort of at least semipublic statement that you will repay, not issue something again, etcetera. In fact, you can think of the double spend problem as just a very special case of this. Double spend is basically like I am giving you this currency and I’m publicly stating that I am not going to give this currency to anyone else in the future and hold me to it.
You know, you’re doing that in a collateralized setting where you just have the currency in your account, but there’s lots of cases where there are properties of you that are valuable but that you wouldn’t sell. All labor is like that, right. Our labor is valuable, but you also don’t sell yourself into slavery, right. So, just the idea of work, anything that is not just a purely financial asset relies on this principle of having properties about yourself that have value to others but that you don’t offer as transferrable collateral in the case that you’re not able to deliver on that, but you’ll accept the reputational hit that you will face as a result.
Again, nothing in this is super creative or new, and there’s lots of other things that could potentially be used as substrates for this, but they have to share this property that there are elements of it that are at least partly public so you’re able to make that commitment that if you break it, people hold you accountable for it. What’s powerful about doing it in a decentralized way is that it opens the possibility that you can do this in a community-based way where people are horizontally lending to each other or guaranteeing to each other that they’ll pay back if another one doesn’t pay back, etcetera, like the Grameen Bank does. Whereas if you do it in a purely top-down way, things tend to be based on sort of a single credit score that is shared across all context rather than on those relational aspects.
Laura Shin:
Yeah. I was feeling that it’s more like a…I don’t know if holistic FICO score is the way but just something more contextual, I guess, rather than the FICO score can only see what it can see. One other thing that I found interesting in your proposal was you talked about how you could use Soulbound NFTs in DAOs in a way to help support the diversity of a DAO. How could you do that?
Puja Ohlhaver:
Yeah. So, if you imagine each member of a DAO having a set of their own memberships and affiliations, right, you could, if you wanted to, say, surface minority voices within the DAO that might be overshadowed by majoritarian influences, you could programmatically bring these voices to the center of the DAO and elevate them. If you wanted the DAO to represent the members of the DAO you could also surface those most intersectional members who kind of best represent the DAO as well. So, there’s a lot of really interesting ways that we can encourage pluralism within DAOs that we really can’t today because we don’t have, really, any visibility beyond just like the wallet, right.
Glen Weyl:
Let me give an analogy. So, there were kind of a couple of paradigms that often compete with each other for thinking about things like college admissions. So one is like meritocracy. The best people should all get in. There’s some score that you’re going to give to everyone, and we’re going to get the best. Another paradigm is like, well, maybe there should just be some amount of randomness to it so that we get something that’s representative of the population. An alternative that I find very appealing is to say, well, sure we want something like meritocracy, but the merit is contextual. It depends on what else is represented, and if we want a student body that is maximally diverse and maximally benefitting from the other people who are on campus then actually it’s not like a select all the top ones problem. It’s a combinatorial optimization problem over a collection of people that will together be diverse, representative, and interact in a positive way. If you want within a Web3 context to solve that type of a problem, you need to think about algorithms that can do combinatorial optimizations over a rich set of things that constitute people and not just how much money does someone or how much tokens does someone have, you know.
Laura Shin:
Which reminds me also that another thing that I found interesting about the paper was that you talked about quadratic funding, which you had discussed previously when you came on my show. I really liked seeing how it’s being applied and some of these DAOs like Gitcoin DAO where, for people who don’t know, Kevin Owocki, the founder of Gitcoin DAO, and maybe he got this phrasing from you. He says he likes to think of it as optimizing the preferences of the many and the poor over those of the few and the rich and essentially if people vote for a grant proposal and they don’t have as much money, it’s like lower dollar amounts that they’re allocating to it but there’s many more of these people and then there’s another proposal where some whales allocate a whole bunch of tokens but really it’s just like two or three entities that actually really support that one. Then the proposal that has many more entities voting for it will get more of the funding.
But what was interesting is in this paper that you guys wrote recently, you talked about kind of the pitfalls of that and said that quadratic funding doesn’t scale but you felt that Soulbound NFTs could help resolve some of the issues there, so can you talk about that?
Puja Ohlhaver:
Sure. So, the problem with just simple quadratic funding is, say, a highly coordinated group could just swamp a mechanism by, say, each giving a dollar, right. Because as you said, you reward the participation of the many over the few. So, what Soulbound Tokens do is they give us a social substrate and let us actually look at what are the weak affiliations and strong solidarities amongst people that lets us discount those groups that may be over-coordinating and trying to swamp the mechanism. In that way, you get back to that principle of cooperation across difference. You get consensus amongst the most diverse members, and you can elevate that with correlation discounts and let the mechanism scale. Glen, it’s your mechanism so maybe you have something to say.
Glen Weyl:
Yeah. I think what I would just say is that, Laura, I actually really love the example you gave of like those big entities because even those big entities will tend to go beyond the issue of a pure civil attack and them just inventing people. They’re going to have a bunch of hangers on of various kinds, you know, like every big entity does. You don’t want the mechanism to just get taken over by like a bunch of hangers on of some centralized entity who just give a tiny amount just because they’re hanging on to whatever the momentum is there, right. You actually want people who are genuinely independent of each other and you want to match those contributions.
There’s just no way to do that, certainly with basic quadratic funding, not even with some sort of proof of humanity because it treats all people as sort of interchangeable. People aren’t interchangeable. People are socially situated, and the problem that we need to solve is not really the problem of getting selfish people to actually lean in the common good but rather the problem of getting people who are partial to certain communities to work with people who are farther away from them socially and that’s what we hope we can get towards with these types of structures.
Laura Shin:
So you’re saying basically that through the Soulbound Tokens, people could say, oh, okay, well, community A, which has this ringleader has been sort of flooding the system with their support but now we see that even though they have strength in numbers that actually when you look at the diversity of groups that make up this organization that communities B, C, and D are also not represented in this vote, even though the number of tokens being allocated to this vote is high or something like that. Is that kind of what you’re getting at?
Glen Weyl:
Exactly. And it comes back to exactly the thing that you said about the different groups in your life that you would appoint for the recovery of your wallet. What you want is not, oh, I need a ton of friends but if all those friends are like crypto bros that you met at the same party, that’s probably not a really good indication that that’s good for you. In fact, that might be precisely a worst-case scenario that they’re like a gang of people who kidnapped you and then took over your social recovery thing or something like that. Whereas if it really is people from completely different parts of your life that’s a much stronger signal.
Fundamentally, again, returning to the point before, this is the right set of logics for a truly networked society where the thing that is at its core is not a set of individuals. It’s not a set of centralized entities. It is the fabric of relationships, the complex fabric of relationships that connects things and actually allowing systems to represent and compute over that. That’s, to me, what decentralized society is all about.
Laura Shin:
So, would Soulbound Tokens then also help mitigate against Sybil attacks, which are when, for instance, one entity creates multiple wallets to get this illusion of representing many people or is that not possible? Because a part of me was like you could just have AIs that are creating these different Souls with these different Soulbound Tokens but maybe not. I don’t know. Maybe it would be hard to do that.
Puja Ohlhaver:
It’s an interesting point and it’s exactly why we recommend implementing this to actually start with humans and actual bona fide human relationships with really rich communication channels because if you did, say, try to bootstrap the system off of just AIs fabricating relationships, then it sort of pollutes all the signals. So, it is better to have Soulbound Tokens actually reflecting memberships. If you have a sort of sufficient level of real sociality, then you can start to differentiate between what are bot wallets and bot relationships because they, frankly, are those relationships that are outside of your social context and are just kind of floating in a network that is very disconnected from yours or even from any other sort of social being.
So, it is really important to start these networks with sociality and not have them overrun by bots. But to answer your question, yes, it could very much help with Sybil resistance because those Souls that have more SBTs that are, say, reputable SBTs can be differentiated from those bot Souls that have just sort of farmed and accumulated fake SBTs.
Laura Shin:
Oh, so it sounds like Sybil attacks will then be possible with Soulbound Tokens as well.
Glen Weyl:
I mean some form of Sybil attack is always going to be possible at some level, right. But I think this is a very powerful mitigation technique for Sybil attacks. So, let’s imagine that people have their LinkedIn, the set of things that are in LinkedIn in their wallet in this world, again to make it very concrete. So, first of all, some set of credentials that you can have in LinkedIn are going to be more recognizable than others to certain people not to others, right. So, it’s not a universal thing. It’s not like these are the good ones and these are the bad ones. There are going to be DAOs that are going to be recognized by some people and other people will recognize the Harvards and JP Morgans of the world and other people will recognize different African American community groups and whatever.
So, it’s deliberately pluralistic, and we don’t think there’s going to be like one like this is a reputable institution and this is not. It’ll be contextual, but in those contexts there are going to be…any organization that’s trying to validate people, maybe some DAO or whatever is probably going to have some whitelisted and maybe some grey-listed set of affiliations that make sense to them, and there are going to be ways of differentiating whether something is the same person or not in most contexts.
So, if someone has a rich CV there and there are two different entities coming along that went to different universities during the same years, now maybe that could be the same person. Maybe they went to two universities at the same time, but it’s not very likely, you know what I mean? There’s usually going to be elements in that track record that are going to be very likely to be inconsistent with that being two people and certainly very inconsistent with it being more than two people.
So, even if you could do a little bit of a Sybil attack it’s going to be hard to do a lot. I think in some ways, more importantly, it also deals with what I would call soft Sybil attacks, which is not that you literally get a fake identity but rather that you get someone who’s completely uninvolved, some complete rando off the street but who just is your friend and you bring them in. That’s going to be much, much harder in this world because you’re going to have that substrate of social relationships and affiliations to either identify a Sybil or even if you don’t identify a Sybil show that this person is so correlated with the other person and the things that I recognize in them are so connected that they should basically be lumped together and not treated as separate entities.
Laura Shin:
So, at the moment, Optimism, which is one of the layer twos on Ethereum has begun releasing the tokens that are going to be part of its governance structure, and one of the aspects of its governance will be that they will have something called The Citizens House, which is where Soulbound Tokens will be used. I don’t think they’ve released too many details, but I wondered, first of all, if you had worked with the team on how they might use them or in general just based on what you’ve seen, what your thoughts are on either how they might be used in Optimism or just in DAOs generally.
Puja Ohlhaver:
I didn’t personally work with Optimism. I think Vitalik probably had some influence there in helping them think through it, and I haven’t looked carefully at their model but I do think it is a first step, but what I would like to see is more protocols and more DAOs having Soulbound Tokens. Once you actually have the sort of plurality of representations of your memberships and affiliations then you can really start to do cool things like we’re talking about like mitigating Sybil attacks and having richer forms of pluralistic governance.
So, I think it is a first step but I haven’t looked at it too carefully to comment on what ways I think it could be improved or not. I don’t know if you have thoughts, Glen.
Glen Weyl:
I don’t think they’ve released enough details for me to get clarity. I’ve talked to people at Optimism over the course of years about this several times but not since the Vitalik paper came out. So, I don’t know what influence happened in what way. But I do know that Gitcoin is heading down a similar route. They’re strongly thinking about using an interaction between Ceramic and the verifiable credential standard to give people various SBT-like things to represent everything from physical location to memberships or contributor ships to different open-source software projects and to start doing these types of analyses. We’ve been in close touch with them. In fact, you’re the second podcast that we’re doing on this topic this week. The first one was with Kevin, and we talked quite a bit about the interactions with the Gitcoin ecosystem on that.
Laura Shin:
Cool. I’ll have to check that episode out. It’s been on my list forever to have Kevin on the show here actually, so I will also have to make that happen. What do you feel needs to happen next for this concept of Soulbound Tokens to kind of really take the next step?
Puja Ohlhaver:
I’m planning a research agenda around this, and I think a research community convening around the different questions, community recovery, primary question and a lot of the incentive compatibility issues we raise around privacy and splitting yourself amongst multiple Souls and gaming, all of this is going to be intersectional research. Glen, you’re doing the same on the academic front. Maybe you want to talk about that.
Glen Weyl:
Yeah. I’m working in a bunch of different ways that will become more public soon trying to create a network of academic researchers to make the study of these types of questions into like a field. The same way that AI or cryptography are fields that can actually do experiments with and tackle some of these questions in collaboration with people that are actually out there experimenting with them in the wild not just the Web3 community but also what’s going on in Taiwan. There’s a lot of interesting stuff there. There are things going on in video games. So, there’s actually a variety of contexts where there are experiments with these going on, and they don’t need to be tied to any particular infrastructure, whether it be VCs or chains. They can be just done on centralized servers as well.
I mean they’re relevant in many contexts, I believe, even though I think they’ll reach a higher potential if they’re grounded in a more complete network community foundation, but I think all those things can be really useful experiments. I agree with Puja that one of the most pressing is these community recovery ones. Another one that I think is really powerful and important is what I would call socially programmable privacy. We discuss this in the paper. It’s the idea that we need to move beyond the notion of what’s often called minimal disclosure where you get a credential and you can reveal about it whatever you want, to one where social groups can specify disclosure procedures.
So, it could be the case that…take like GitHub for example, or something. It might be that I get a credential but I am not allowed to disclose that I am a contributor without also disclosing the names of the other contributors because that is credit sharing and whatever. Or you might want it to be the case that before disclosing something, I need to check in with someone, right. So, I actually think what we want, because it reveals information about them. What we actually want or what I want is not privacy in the individual I can reveal whatever I want or not way but rather that social groups can together co-determine what’s revealed to whom and how.
Laura Shin:
All right. Well, this has been a really interesting discussion. I think we’ll sort of have to see how things play out over the next year or so because as we mentioned I think people are going to be experimenting a lot with this concept. So, where can people learn more about each of you and your work?
Puja Ohlhaver:
I guess Twitter. I don’t really have a public research page, but I work at Flashbots and we’re thinking about the problem of MEV, and I see that as a special case around what Glen just talked about, programmable privacy and moving away from…if you think about it, badge auctions are kind of the sort of collective group data cooperatives that Glen is referencing. So, if we can solve this problem with atomistic transactions in Ethereum, we actually can solve that in a lot of other more use cases. In some sense, it’s actually probably the hardest on-chain problem to solve from the perspective of programmable privacy. So, that is what I spend my time doing. Anyone’s free to reach out to me on Twitter. Always happy to talk about research.
Laura Shin:
Yeah. Can you just give your Twitter handle?
Puja Ohlhaver:
@PujaOhlhaver.
Glen Weyl:
And you can follow me on Twitter @GlenWeyl, Radical Exchange website and blog carry a lot of materials I produce and I also have a website, GlenWeyl.com and SSRN, so there’s a variety of places to interact with my work.
Laura Shin:
Perfect. Well, thank you both so much for coming on Unchained.
Glen Weyl:
Thank you, Laura.
Puja Ohlhaver:
Thank you for having us.
Laura Shin:
Thanks so much for joining us today. To learn more about Soulbound Tokens, Glen, and Puja check out the show notes for this episode. Unchained is produced by me, Laura Shin with help from Anthony Yoon, Matt Pilchard, Mark Murdock, Shashank, and CLK Transcription. Thanks for listening.