MEV From the Dark Forest to institutional adoption - Samuele Rodi

Transcript

thank you thank you thank you it's great to be here and uh it's great also to see like a many Survival on a Friday afternoon actually after so many talks and uh like intensive talks as well particularly um so as you guess from the title uh today I'm going to talk a little bit about some scary stories uh more or less let's say so uh let's dive right into the topic as was already being introduced and I want to start with this uh ethereum is a dark Forest so even though the term I was Co back six more than six years ago and

earlier than this it actually became very popular only with this article which was like written by the guys at parim and uh I invite you all to read it because like super interesting and uh super exciting even though we're talking about like you know some years ago still proof of work on ethereum so completely different landcape and actually what's happening in this article is that the the comparison of ethereum to a dark Forest which uh for those of you who don't know the Dark Forest is the hypothesis that Advan technologically advanced civilization actually tries to hide themsel uh in order to prevent uh in order to prevent the detection by

other partis and so so that they can guarantee their survival and long story short of the article was basically um the article was talking about an attempt to recover some funds on a on a protocol it was discovered vulnerable but the authors were really aware of what was happening in the mle back then so there was like mostly just public mols back then and so every transaction was visible and uh so they tried to hide this transaction uh be with the goal to prevent being front run by other Bots long story short that atem was not successful and so we

got a lesson which is actually ah yeah it is there so don't rely on normal infrastructure so after this article uh there were many follow-ups many debates about how we actually achieve uh an ecosystem which is protected from this kind of threat and we can prevent like front running because remember you and it is not just about Arbitrage as many people talk but it's like a lot many other things it's liquidation it's front Runing and so on and so many people talk about like private mles so introduce fure of mles uh in some certain Min pool back then but it was only until these guys came into

picture that uh they actually set up a clear path forward and I will skip very quickly through the the story of it but basically F very fast forward to today we now have a PBS and it's still the flash BS guide that are basically leading the the way toward the the vision of of the ecosystem and what is PBS for those of you doesn't know PBS stands for a proposer Builder separation and it's a concept in ethereum that aims at separating the proposers which are the valid s uh from the actual Builders and you see this is the complete

architecture um of the PBS ecosystem which is now implemented via M boost which is a software developed by the spash guide so to explain you how it works very quickly a regular user normally send a transaction to the public mempool and now there are still like these uh Bots MV Bots around there which are called like searchers but actually look for this transaction and still try to sometimes front FR you so this is still the case this still happening and what happens is that the end they send this transaction to these entities that are called Builders and they are the one that run very expensive machines and they are responsible to

basically build blocks for the entire the ecosystem and there are few of them obviously it's not just one but uh not so many and once they have like this block produced they actually try to produce a block which is optimized in terms of profits they try to maximize the profit for both the builders on selves and the validators so then they send the block through some relays and the relays bya me boost which is a sidecar software of the validat to client everything gets included in chain and so here in this picture also very important the relays are basically the trusted intermediaries of the

architecture so basically they guarantee that the Builder are producing valid blocks they validate the blocks and they also make sure that the validators are not stealing those blocks so they're really The Trusted piece in the architecture and so what happened for so why am I telling you this it's because like nowadays with this Builder setup we also have like you see there there's like uh New Concept of private order flow which is like really really powerful and not I think not everyone is still aware of so that's why I'm I'm actually having this presentation and let's say a user which is aware of this MV threat actually can use

um can send bundles of transaction directly to to this Builder which is called also referred to as a builder RPC and in this situation so basically we are bypassing completely the mol we we are just sending transaction to the builders which by keep them private and send them to U the chain so now here comes one of the scary story and I might ask you here uh to all of you so uh come on raise your hand like what do you think is more decentralized uh ethereum do you think is ethereum more decentralized than Bitcoin come on raise everyone the end

there you go okay so Bitcoin is still more centralized we look at some charts so the very good thing about this new PBS architecture is in that you see on the on your left uh PBS architecture was very um much used since the proof of uh uh proof of stake introduction since the merge and need like all the most more than 90% of the block were actually built through this PBS new architecture because there are still some percentage of block that are built through a normal uh vanilla Builder but

they try to build block themselves it's just like solo stakers running the whole thing themselves and the Rel scenario so it's pretty diverse so there are as I was saying before there are like a few Rel layers but it's it's quite fragmented what is instead like a little bit scary concerning is the other chart on the on the right there you see actually the the share of the builders that are actually building blocks on ethereum and um I guess you can read the that pretty well like if you zoom in at nowadays there are basically two block Builders dominating the the entire the

ecosystem and there's good and bad like these guys are running very powerful software and the entire PBS architecture was set up because of this because they wanted to have uh Builders running powerful infrastructure and and everyone instead like running validators being able to do it at home obviously this is not possible for everyone to run such complex infrastructure so this is the tendency and it's particularly concerning because I tell you one of the two is actually sensor transaction and uh one more thing they're actually engaging we've seen other talks so it's not a New Concept because a lot of people already talk

about this and um these these Builders sometimes are actually building blocks by subsidizing the blocks meaning they they're actually producing blocks at their own losses with the goal of actually increasing their share because more blocks produce people Searchers go to them and so the more sh the they have the more sh the gain and so on and but anyway this was just one short story and Side Story look let's look at some uh nicer metrics user side Builder RPC so we've seen let as side like glitch in the chart because these are the very

difficult metrics to collect um but we've seen an increase in um utilization of private transaction meaning like those trans that funnel through this PBS schema and if you were wondering about inclusion time which is like if you have understood how the architecture looks like it is might be the most concerning part for a user uh actually we see inclusion time is really is really not a problem uh normally transaction gets included within at most two blocks but so you might be asking why am I telling you all of this so I just want to give you a perspective from um what

the institutions are looking at and also like in particular as as Signum bank so we've said actually there are good things about uh this Builder rpcs and these are the form basic ingredients which are very powerful so we're talking about Bandos meaning the possibility of sending Atomic Bandos uh transaction in where gets included within a a single block talking about possibility of canceling transaction for free on chain which is something that was not existent for the modex sa view you might know that to cancel a transaction on ethereum you actually have to send another transaction invalidating the previous

one here you have it for free and it also comes with you can also expire transaction for example another important thing is reverse protection meaning uh if a transaction is meant to fail on chain by using this Builder RPC it's not goingon to get included in the block and the last thing is privacy yes we talk about it and which means until the transaction is included no one except the Builder has the possibility to see it and so we have our basic ingredients let's make our pizza with those so what can we do with this like for example let's stay on the trival side of things very simple uh staff ux

Improvement so and this is particularly important because when I talk about institution I talk about entities with let's say high grade security and some sort of automation that I can actually operate on on wallets in this case and ux improvements are particularly interesting because you can now with this setup uh execute operating on on a protocol executing an action in one single action instead of I guess you know by interacting with metamask when you interact with a protocol you have to always approve wait send and then click buy or sell or maybe approve multiple

times super annoying very natural sponsor transaction so now we can that we can create a bundle we also can create a bundle with transaction of different nature so we can actually have one sponsor wallet to pay for uh the transaction fees for for another executor so that basically we can have like wallets that trade or transfer like us 20s whatever you us Toc without having G without having to worry about ethereum at all and this is actually particularly important for us and something we are proud of and why I'm saying this uh because if you guys don't know what Signum bank is

doing we are very known for our offering which is our one part of one our core value proposition which are um segregated wallets what does it mean that each client has its own wallet and they can see it clearly on chain so very simple and actually like clients love it because of its simp simplicity so because they canot go on chain and they're sure that funds are not commingled with each other and that's why so this obviously is a very good feature but also comes at a cost for us because we um we have to manage a lot of wallets for all the clients and we have to come up with some

efficient mechanism to actually manage those wallets so that's why we're really interested in this let's say offchain feature which come with for example the Builder rpcs that I'm describing now and um yeah so I guess it's something to look on in my opinion whatever is being built off chain nowadays is actually becoming a lot a lot more important and fundamental that whatever is happening onchain because onchain is being build very slow uh while offchain is very very Dynamic so keep an eye on that moving forward limit orders of the amm I know that some amm have the capability of you

of defining limit orders but with this set up you can basically do limit orders without actually having to worry about protocols specific features because you can send a transaction making sure it only execute upon certain condition and if not it won't be included so you keep on retrying until your order goes through last but not least instance the settlement so in the same way with B transaction we also can interact with different protocols at the same time trade on one side um buy on one side sell on the other at the same time no risk in uh in changing ing rates so we are really I'm just showing you that we can reduce a lot the risk of interacting

with onchain protocols by using these features okay this looks different than expected but anyway uh I recap it very quickly for you uh we looked at some technical Matrix around this Builder RPC and uh we just assessed transaction speed which is the four stat I guess you don't see it and we basically understood that we saw it before it depends mostly on the network share in in this setup but we're we're we're confident that by selecting appropriately the builders you want to R your transaction through the transaction can go in fairly quickly in terms of

scalability Builders are really not the bottleneck here it's a AVM because Builders are super powerful machine and cost the four point is cost batching can actually reduce cost because you can run strategies that can be optimized while if you just submit uh transaction to the mol and you just wait for their inclusion you might not know what actually you're spending here instead you can you can predict it well ahead in terms of reliability again these Builders are designed for profit maximization they will be up and last but not least it's privacy one is private by Design the

other is not so we come to the conclusion here and so one thing that we can say is that this PBS architecture at least from an Institutional perspective is very robust and stable we see there is indeed an adoption an increasing adoption of private transaction which is also on one side a good sign uh this Builder RPC unlocked very new use cases building of chain as I said stress out again generally yields faster result and with much lower complexity like think about RC 4337 for account obstruction it's still not very clear where it's going to be

honest obviously there are some open points there are some con to think of because like builders in this setup are kind of centralized entity uh but there's no legal contract in place so this is very concerning from institution perspective but we believe that with the right setup and with the right risk mitigation procedure we can actually accept the risk and this is a concern not for institution but for ethereum foundation that this Builder centralization is a much bigger concern practically than whether validate or decentralization but because like a lot of time they talk about the centralizing inv validators

but we've seen we might have like some bigger problems on the Builder side so there are more stuff coming on very soon so unfortunately didn't have time to talk about uh all the Innovation uh so epbs or Suave it's just that all the stuff is still being discussed and progress is like moving quite slowly in general and also we are aware of Spector grade which actually finally bring on chain some of the feature I discussed about but just to refrain that uh from institution institutional perspective this is what is the standard now what is solid and fully functional

and so basically as institution we normally rely on software that's being running being battle tested and being up for a long time so anyway watch out also for those upgrades and uh with that I conclude my presentation so thank you very much uh Samuel I think we have time maybe for one question yeah so uh our clients know we um we have externally own accounts for as if you refer to segregated wallets we do have Smart contracts as part of our

offering but it's doesn't um there's nothing to do with what I describe today yes yeah right we actually have one smart contract for that purpose yeah but it's not a client's account in that case so uh thank you we'll take just one more question and then we will close so um your presentation was awesome thank you thank you but I think everyone should be shocked and horrified by what he actually said um just to summarize um uh sophisticated Bots are

stealing money from normal users and the solution was that we don't rely on normal infrastructure that's horrible the solution is that we have this outof protocol solution where we introduce trusted intermediaries and that leads to more centralization and things like payment for order flow these go against everything we stand for in ethereum um I just want to say that I'm here presenting facts not opinions my question for you is since you presented the Band-Aid approach an outof protocol solution what might be an in protocol solution to the problem of

malicious me uh um so I mean there's this epbs concept which I unfortunately didn't have time to go through and I mean I am not part of the discussion in ethereum foundation so I'm not not actually 100% sure of the evolution there but yes I see there are some good thoughts around it and then the idea is basically to uh propose a block which actually gets in um included within a c with a certain delay that for example can bring some benefit it and U but again like the discussion is still really ongoing it's still very early and for me the important Point here is that just a call

to action that we should think more and more about uh these problems and try to address it cool thank you very much yeah thanks for the question and that reminds me also the discussion we had before on the the panel about the spectrum of decentralization right so it's not black or white so it's important to yeah to discuss it so thanks a lot Sam I think we we learned a lot thank you [Music]