The Quorum Crisis

Transcript

we're facing very similar problems and I hope to be able to make that clear about how do we see that happening uh during this talk. So as I said I'm Zoo. I'm head of research at BlockFool. Uh we built anti-apture that is uh governance security framework and dashboard. Uh the goal is to make Dows safer. And we see coordination as our goal. We see governance as the path to coordinate better. And for a long time we've looked at Dows as a great experiment to learn more about governance.

The truth is dows are now too big for us to just think this experiment is allowed to fail by lack of attention, by lack of care. Uh so we're doing our best to move that forward in a safe way. We got to building anti-apture after a few years working in governance through different fronts. We've explored from reputation voting systems to escape this plutoaucratic uh totality that most of our Dows are plutoaucratic today. So experimenting with different ways to do uh governance.

We've explored AIS for for delegation for voting and we've been working as contributors, service providers, stewards, delegates in a few dials over the last few years and we got to the conclusion that security was being underlooked that security needed more attention and decided to take it in our own hands. And this happened in this this year mostly what we've been seeing as the year of the Dow challenges. We've had a lot of hype over the past few years. Is this the year of the Dow 2022?

Everybody's like, "Yeah, we got to take over the world." And things change and bare markets come. And now in 2025, we see lowering turnouts, less and less people taking part in governance. We see what we're calling a governance blockade starting to happen in some Dows where freezing the DAO making it nonoperation oper operational is a new challenge and we also see vote buying and bribing starting to go around in different ways and those are some of the things we're qualifying as symptoms of the same quorum crisis. Uh one of those one of the big examples for this is recently

unis swap uh having a big challenge to reach quorum. Uh unis swap has been running as a dowel for a long time and now in 2025 they are just barely getting over the margin of the minimum amount of votes they need to approve anything. And when I say to approve anything this is a livveness check on the DA. If you don't get to quorum, if you're not able to get enough people for any proposal to pass, that is just a frozen address. Nothing is gonna happen there. And that's been great work on making this move out of the problem. We've seen uh the work in unis swap as one of the most interesting works on fighting this quorum crisis.

But we also see this quorum crisis in many other places. Arbitum is another of the big examples that is facing this challenge. Arbitum has a little bit of a more complex quorum [snorts] system. It's not a hard forever number like unis swap has now. Uh and making sure we keep reaching that quorum on arbitum and in a manner defines the fate of one of the biggest layer 2s that we have right now. We're talking about billions in TVL uh that need to elect security council to be able to upgrade and to advance the protocol itself on the other chains on if you're in mainet mainly but arbitum one arbitum nova is all

dependent on this being possible and I love the thread there because those questions are exactly what I think we should be looking at why are delegates losing voting power why is it hard to gain new voting power and what can be done to tackle this and what's the main reason but the quorum crisis doesn't express itself only directly in quorum the quorum crisis also expresses itself through whale captures as we have in compound now that for more than a year now has humpy as the lead of this big coalition that currently holds 33% of the votable supply the delegated

supply has 33% is 32% in a single coalition that last year tried a few different mechanics for attack. This year was able to block uh the approval of a budget for a security council. Tried to decrease the treasury delegations to delegates. It is it is a game a balance of powers but this is not 33% of the total supply. This is 33% of what's delegated because the quorum crisis expresses itself a lot through the lack of delegation and the lack inside the delegation the lack of turnout. So we can see that Humpy now has more voting power in his coalition

that we know of because most of those wallets are anonymous wallets coming from centralized exchanges that we can't really map everything. But what we know is already bigger than the average turnout in the DAO. And one more way this all expresses is when we look into the the consequence you can have if the quorum crisis and the lack of delegations uh takes shape without care. an ENS. We have a treasury not counting ENS tokens only Ethereum and stable coins that is almost three times now as big as the total delegated supply. So that means that if there was

no security council on ENS that we've put in place together last year right now that's a like assuming every single delegated token votes that's still quite some profit to make that's still a very good two threex for you to get if you're an arbitrageer very well funded or if you're competing with ENS and you want to make the DAO go down and you have incentives gives to to freeze it to destroy it. That would be a vector of attack that's still profitable. On top of it, there's also vote buying and yes, recently we acted as blockful on arbitum

uh to buy uh votes during the security council election to show that vote buying is something that you need to pay attention to. This was an issue that was exposed for a while in arbitrum and not well taken care of. Uh gladly we got our desired result that was arbitrum delegates uh were able to repeal our our tentative of attack to get into the security council and the defense was well succeeded. But we did manage to get 14 million votes. Almost enough to get you in the security council in like way way way cheaper than what this could count could cost, which

would never be enough if delegations are bigger, if quorum was higher and all of those things. Oh, you can read more if you're interested on this specific case of of what we did in arbitrum. Now, uh you can check it on paragraph on our paragraph. uh all of the articles I'm going to mention uh are there and well it was interesting to see the defense being able to push us out even though we're going to like anyway pass the seat but it's good to see that arbitrum could defend itself from that but all of this is kind of old news right I don't think anyone here is like wow what do you mean dows are under quorum like no we know that and we tend to just

look and say that's how it goes But this is called normaly bias. And normaly bias is a dangerous thing. Normaly bias make us think that we're safe, that it's not going to happen to us, that our case is different. And many times it's not. Many times is a matter of chance of time. And we need to look into it by understanding the the structure. For anyone not too familiar, I'm going to quickly go to the compound governor. It is like how most of the Dows that have representants here or that we are working with are established. It's basically a token contract. It's basically what you hold.

The governor is where you vote and the time lock is the treasury that executes that. The way that it flows through is that you can buy and delegate in the token contract. You can propose and vote with enough power in the governor and given the rules that are set the time lock will execute that. I guess most people here are already familiar but always good to just brush back and was looking into that structure that the first moment we saw quorum as an issue was on ENS and we found the first signs by looking at how much of that supply was actively governing the DAO. Back last year 4% total was delegated and we were like okay what does that mean? What risk does that

offer? In that case you could see that the average quorum the average turnout was $125 million and the treasury was sitting at 120. That's very good five four almost 5x. That's like that's a good arbitrage for you to make if you want to make some money. And well, that's not the most of how it could get exploited. The average quorum could have other mechanisms to play together with an attacker here and exploit it even more through things like beyond the economic conditions through parts of the governance implementation itself. All of those governor contracts, they're normally customized. There's no no one's

running vanilla the same everywhere. And in this customizations and in the configurations, you have things like what's the voting delay? What's the time between my proposal and when votes get counted and voting starts? ENS has one block voting delay. That means that 12 seconds after I propose, voting can already start running. That is very little reaction time for us to have. and those types of things. Finding those issues is how we protect our Dows from the quorum crisis as we work to revert it. As we work to bring people back, this is the p this type of thing is a possible path to make it harder to extract. In the ENS case, we saw that a

the governance stack had the a few of those weak configurations and in a bad economic condition, it could turn to an easy attack. it could turn to a capture to one of the most important and respected protocols in the space and well God knows what that would do to Ethereum reputation in general. Uh that's also something you can read more and explore in our paragraph if you want but I want to go further because we saw the issue wasn't just there. We started digging. We looked into more than 30 attacks and we started to find patterns.

There are patterns of how those attacks become possible. There are patterns of what the economic condition looks like or what the governance implementation looks like. And we got out of that with four verticals to analyze how safe the Dow is at any point. First one is economic condition. Is it profitable to attack? If I'm going to spend 50 million to get 10, I'm not going to do it. If I'm going to spend 50 to get a 100, I might token distribution and where those tokens sit. So, who has the power and how can I get those tokens? Is this all in a centralized exchange? Can I OTC 60% of your voting power? That's doesn't sound comfortable. That's a way where

you get more exposed governance implementation. So, inside those contracts, what are the rules that allow me to make it even cheaper to attack? So the economic condition doesn't need to be so favorable and I still get to profit and governance activity the core point of the quantum crisis that is how many tokens are actually being used to govern. We see that three of those points are related and are overexposed by the quorum crisis. economic condition, token distribution, and governance activity are all part of the lack of participation we have right now of the amount of token holders that delegated years ago to someone who doesn't vote anymore to the amount of tokens that were bought in uh AMM and

have never been delegated after. And this is all sitting power waiting for someone to come and grasp. Yeah, it was the year of the Dow challenges, but it was also the year of looking for solutions and solutions that were worked through by people on many of teams that are showing up here that we're very glad to have worked with or observed the work into ways to fight this. As I said, this is not news and we've seen goods good and bad solutions in our perspectives. Uh I think quarum reductions were discussed in many places. uh it that's a dangerous one. Optimistic proposals were put uh in a few places too and those are

quite interesting ways to make work easier inside the governance. There incentives campaign treasury delegations there are many approaches that we've been seeing around and uh better or worse there are search for how to solve this. So they're a good signal that we are recognizing that we have similar problems. But we face a challenge at that. The similar problems can be solved in a few different ways. And if we look into keeping our organization safe, how do we do that without surrendering to trustbased systems? How do we make organizations stay safe without saying like, okay, there's going to be those three guys and they're actually going to

call the shots. They're pretty cool. You can trust them. It's not the path uh we would like to to look for. Getting holders to take part in governance. Getting holders to care and understanding that's important part of the value they hold is what we think is is the path ahead. And I think that resonated a lot uh when Yo, Vitalik and Marisa Posner posted the trustless manifesto this week. Uh if you haven't read that yet, I really recommend you do. This is quite a good piece to see in the middle of institutional adoption era of Ethereion. Seeing that there's a stance and there's a some clarity of how to to progress in the same ethos and alignment that have got us here. Uh and

if we want to have decentralized governance as one of the things that keeps our system safe, we need to make the decentralized govern itself safe so we're able to do that to make it trustless to make it resilient. And in this search for improvements, LDG, the Lidos dual governance was one of my favorite advances of this year in improving how we keep things safe uh while keeping it trustless. LDG on Lido lets LDL holders vote and govern, but lets the users with state if uh balance the power of each other. So

state if can freeze governance, can block proposals, can stop the protocol while LGO makes the choices of how things should move. It's one governance made by two systems that currency balances like 20 billion or something in TVL and it seems to be working. Uh not to say Lido is perfect. There's security improvements to touch mostly everywhere we look in this space. But that was one of the most exciting improvements we saw this year. The other path we see that's not as technical and complex as this big diagram but in practice it is harder is to find resilience in the holders to

understand that who owns your protocol who owns this gets to to govern. This is arbitrum total supply of governance. This all it's excluded. This is not counting on governance. And out of everything else, this thin strip here in the middle, this is delegated tokens. This is votable supply. Everything else from there is not touching governance at all. Can you imagine how much harder it would be to get 50% and approve a proposal if that thing was actually the blue one? If we could look at this and say like, oh, the blue one is the delegation and voting. That would be beautiful and way more resilient. But we're not there yet. This is a piece of the trustless manifesto that I think

sinks very well at this that we we don't want to depend on intermediaries that cannot be replaced. We want a way to do this that we're not concentrating trust to a few operators only. And to bring all of those holders, we need to understand a little bit of what's the the issue here, right? You find a protocol that you believe in and you're really fun of it. you think, well, I might as well help govern this thing I care about. It matters to me. I put my capital then in some of the Dow tokens and I'm like, okay, now I'm in.

Let's go to the forum and wow, let me get out of the forum. This is overly complex. There's a lot of context needed. I'm not grasping this. Uh, I have a lot of things to do. Discord is even worse. Telegram groups are hidden. I don't know where to go. I'm gonna delegate to someone I believe. And I did that like two years ago. and never followed. Now the guy hasn't voted for one year in a bit and I go look there and I'm like oh my power is not even being voted on like well at least I wish I could do something with this and then I have an idea I'm just going to put it as LP somewhere get some yud or sell it for something that gives me yud and that

sits on a centralized exchange now ready for the grasp of whoever wants to attack. I think this challenge of bringing holders inside of the system uh is one of the parts that we really need to tackle here. Like we cannot keep hoping for people to just be [snorts] nice people that are willing to lose all the you all the money they could make by voting in votes that they don't really have a voice that actually in the end four or five or maybe 10 15 if you're very lucky and well organized in the DAO are the votes that are actually going to define it. If a few can make a decision we're just lucky that they're not abusing that power. We cannot trust this

assumption. This is not math. This is people. You can see like on both ENS or unis swap those are the number of delegates you'd need to reach quorum. This is a very small number of people giving all the holders we have. And if you look at uni, uni is struggling to reach quorum which means that with top four already reaching quorum that is very likely to be able to pass. If you put top seven unlikely to be defeated. Not only that, a lot of power sits on on centralized exchanges. You have both on ENS and in scroll uh supply on SAX way bigger than the delegated one.

That means that just like it was done in compound by Humpy last year, those tokens can just get in to a lot of new unknown wallets. and be turned into invisible power ready to be used. $10,000 in 10,000 wallets is a lot of voting power. And if they're coming with different amounts at different times from a centralized exchange, it's very hard for you to catch that before it turns into a vote. Not only that, but power that is delegated is also not used. This is unis swap and this is compound and those are their top delegates.

Uh you this might be a little bit outdated. You can find the new ones on anti capture but it's like if you look at unis swap this is the top 10. We have two votes out of the last 18 proposals. Oh three one more down here. Top delegates are not voting. If you're delegating to one of those guys please delegate to someone else. We have unis swap has competent delegates looking into things and doing work. Why is this power all stopped here on compound? We see like Pigov, Evan Horizon, all the callers and Humpy the leader of the attacking coalition. Those are the only very active ones that you will see.

Everything else is stopped. We need to move this move this voting power. not only bring the holders in but move holders that already delegate into useful delegates and to people who are actually making the Dow safer and oper operational those conditions the quorum crisis it leaves us exposed they all leave us exposed and we have to assume Lazarus knows if we know you have to assume Lazarus knows right if you're not aware Lazarus North Korean attacking group, hacker group, billions of dollars on buy bit and yada yada and well they like to make money when we make

mistakes. If we see the risk before the problem like you have on compound and ENS where you can put a security console in place you can guarantee a veto there that's nice but you also have to guarantee you can renew that right having a security council like compound has until September I'm really hoping that compound that the foundation the groups working with them can move it forward to change this scenario before the renewal of security council comes because the risk is given compound Dow very real DAO very much the real thing it has control over all the protocol you do not want that in the wrong hands with like over 50%.

>> [snorts] >> We have to assume Lazarus knows and know that they are creative. They're smart and they're well capitalized. Normaly bias is art trouble. We have just assumed that nothing is happening. We have data. This is all open public blockchain. Yay. But we don't look at it. And if we don't look at it, we don't know how to play with it. We think 51% of the delegated supply is the only way to attack us. And then suddenly the guy is also seeing that this is not so profitable. So he gets it. But then he shorts your token. He gets a loan.

There's many ways to start getting closer and closer to that attack power and making it financially viable. We have to assume Lazarus knows. So we have to make sure that we do too. Cuz if we don't, there might not be many of those Dows left soon. Those treasuries are going to be very dry. And in that capture, we also been looking for solutions. And they're not silver bullets. They're not going to solve everything. There's many other solutions around. And we hope to be able to bring a little bit of our research and what we've been seeing that works into well for all of you to see how we can move it forward together. First part

is evaluating economic and systemic risk. So anticapture is based on 17 different metrics of implementation, a lot of economic data to get to the point where we can tell different risk areas of types of attack and stages of resilience. Uh very much inspired by L2B and how it moves forward uh L2s to have pressure and understand you have to evolve this stage is not reasonable. You're not you need to keep moving notification systems and informing holders and delegates. We all seen a bot that tells you, hey, there's a new proposal, but you also need to know like, hey, there is a new proposal that your delegate hasn't voted. In the last three proposals, the guys are delegating

never voted or this guy that you think might be an attacker has been increasing voting power every week. Being able to have that clarity in the hands of security teams, governance specialists, professional delegates inside the Dows might be very powerful a tool for us. And also incentivizing active delegators. That's one thing that we haven't seen a lot around in in Dows in all those solutions. We've seen people vote a lot in favor of incentivizing delegates that are voting and that's important. I think we need to keep that going. But we also need to make sure that people who are delegating to those

active delegates. Those people they can they should be getting something here too. We need to make sure that they keep doing that. We need to make sure that that increases. It's not a matter of oh we'll are we going to spend too much at that. If we don't move this number forward we're might not have a lot left anyway. depending on how that goes. So that's a few of the paths we've been looking to to improve and challenge the problem and we do that based on a framework the anti-apture framework as I said before we're looking to a lot of different metrics so far we're focused on the compound governor uh open zeppling governor bravo however you want to call

it and we're very much open to hear from people who know about governance what do you think of the way we're looking at this how we expand this further how do we look into other governance models One of the great feedbacks we had so far is that we make we need to generalize that governance is not only governor. There are many ways and we need to make it broader. Recently that is what got us uh seeing the trustless manifesto was Vitalic mentioning that L2B wallet beat and anti-capture have been doing uh ways to effectively measures a system trustlessness. This is what we're going towards here. Understanding that this is like this is working. How do I how do we

know if it's working or not? We get all the data we can and we measure the results of different actions we can take. It's dangerous out there. Please take this as uh as a suggestion of ways that we can all improve. We need to get this space to improve in how governance security is done. Anti-apture looks into a few Dows in stages of security risk areas. We're looking into each of them into a lot of different metrics that you can explore just out there and see on your favorite Dows how it's looking. And we're looking into like the details of like each type of supply, how is it moving, how does it

relate to each proposal, where does each delegate voting power comes from, and when did it get there? like it's looking into these details that allows us to see for example when Humpy acts on compound and it starts accumulating into new addresses and we can see a bunch of new addresses that are getting a lot of voting power on the same day from the same source and now we can flag it which when we saw the attack one year ago we couldn't and want to make sure that everyone can get that type of information and every single D is protected from that type of vector governance was never easy to do all over history. And this is one of the last parts of the trustless manifesto. We choose to build trustless systems even

when it's harder. This is not easy. This is not going to be easy, but you're all here because you care. And a bunch of you I know personally and are very good at this. So, let's make it better. Thank you. [applause] Do we have time for questions? >> Yeah. >> Yeah. >> Okay. Anyone back there?

>> That was awesome. Thank you so much. Um I I have a little bit of a heretical question which is a lot of the problems that you outlined are uh predicated on uh anyone being able to vote and I'm wondering if you've thought about alternative governance systems that don't necessarily allow anyone to buy tokens and therefore participate in voting. I think there's part of the permissionless nature of anyone can vote uh that I appreciate. Uh I think if we don't assume that the way we look at this type of analysis is very different.

I really love the way the way you guys are tackling it with hats uh for the DAO. I find it's a that that does make it safer. But reality is $16 billion between treasuries and over 30 in TVL that are managed like this right now. And we need to make sure that stands before taking next steps. But I'd love to get this type of thing one day to be able to also analyze uh more diverse governance types. Hey, thanks for the talk. Juan from sky here. Sorry about my voice. Um, we've

seen some when we were measuring activity a few seasons ago on the delegates. We've seen a lot of um fake activity or people just almost randomly voting. Well, it wasn't even random. It was like following the trend and just rehashing what the other main delegates were saying. So we ended up with the activity which was a multiplier for compensation and it ended up being like everything read the same. So I don't know if you guys have thought of a solution for that. Well that farming farming voter rewards is something that we see a lot in the space. Uh I don't most of our solutions regarding incentives are focused on the holders

that are delegating to those voters. So not not a lot haven't fought a lot of how to prevent that. The way we we structured our latest uh request for comment on ENS is that you get rewarded proportional to your voting power. Uh doesn't matter how you vote, where you vote, what you say or not because a spammer is going to spam. like you post it on the forum is like it it's important if you post it well but then that's a lot of extra work and not something you can just automate easily.

Uh but yeah that's all this subjective part of what is actual contribution is a whole other layer on top of this. >> You mentioned security councils but another way to prevent governance attacks are proposer councils. Uh what are the pros and cons of those different approaches? >> Proposer councils like a council that holds the power to propose just by themselves is that can can he has the microphone? Can he has gosh the web tree nerd shows sometimes.

>> So a proposer council basically has the responsibility to post any valid proposal uh and put it to the full DAO. But a proposer council can filter out malicious proposals and can also filter out spam attacks. >> Okay, so that's similar to what optimism has uh nowadays, right? The foundation is the only proposer. They're moving forward to change that a bit, but right now optimism keeps that structure and that kind of does it. That kind of holds it in place. You're not going to get a proposal that drains everything you have. You are a foundation which means you might be super node. It means you might freeze the DAO because no one can ever propose again because now proposing

is a crime in 20 different nations and blah blah blah. That's it's like it's not a a fully resilient trustless system. Uh which is what I like to aim towards. But if you cannot protect it any other and you have a trusted body that does keep you keep you safe. I think the way optimism is progressing from that to decentralizing gradually is being well done, but it's not one of the designs I'd see as more let's say uh more aligned with an unstoppable machine. It's aligned with a business.

Uh guys, I think our time is over. Thank you for the questions and thank you for the attention.