Safe Token Launch, EIP 3074, and Technical Updates with co-founder Richard Meissner

Transcript

today's a very special day for the defi ecosystem one of the core players safe the multi this secures hundreds of billions of dollars of crypto assets has announced that their token is going to be transferable in today's video we discuss what this means for safe users and the importance of smart accounts going forward in crypto there's been a lot of talk about EIP 3074 as well as safe's new key store rollups in today's video with co-founder Richard Rob and I dive deep into what this means from a technical architecture standpoint and also how the safe token will play a part in the future hope you enjoy thanks welcome back to the rollup today we are here with Richard one of

the co-founders of safe today the safe token has been announced as being transferable so a lot of work has been put in by the team to create basically the Bedrock of all of defa in terms of asset management and and self- custody so yeah Rob we have the pleasure of sitting down with Richard I know that we did an episode with with Lucas the other co-founder maybe three four years ago um so glad glad to be back yeah it's a it's a testament to safe's longstanding history uh and uh fundamental presence in Defi ethereum and and blockchain space in general so yeah like Andy said uh episode with Lucas was quite a few

years ago a lot has happened since then um and there's still a lot more uh that is going to come out over the over the coming weeks and months here um and so we anticipate that uh the nature of custody will start to change uh and this new paradigm of account abstraction will start to take shape and so I think that's probably uh a good place to start is just defining account abstraction maybe defining like the problem statement here and then we can start to dive into some of the uh proposed eips and and therefore proposed Solutions sure so first of all happy to be here um what is account abstraction um I think it's there are

many ways to it but looking at a technical level right like if you look at especially ethereum and other blockchains how accounts are defined as really this very strict um definition of okay you have a cryptographic key it's a key pair you have a public private key and it's even defined down on the curve like what curve are you using and while you have some um freedom on okay how do you utilize this curve with NPCs and others it's in this case it's quite strictly defined and if you want to um be more flexible we have to do major changes and here it's where account abstraction comes into play right like so um basically we want to abstract away this definition in etherum currently what we have is that we say we have ecdsa curve which is based on SE P um K1

and you have a increasingly monotonous non switch depending on your actions like executing transaction deploying a contract that increases and both of these you can basically open up where say okay actually we do not care about your cryptographic curve we just say you have a smart contract and you can call into the smart contract and this smart contract can however they want verify a signature and we even go one step further we can say however what replay mechanism you want to enforce you enforce it yourself right like um where the protocol normally enforces the monotonous nons and these are the very core Basics um that you will always have when it comes to verification of your account but um especially on the blockchains obviously you also have the

fee payment part which is another level to it where you say hey normally you have to pay in your native coin nearly all blockchains there are some exceptions so um but even there with account OB section you could even go take the step and say yeah it's actually not necessary to pay in your native coin you just have to pay somehow so that some validator or miners accept your transaction your payment right and if we take all of this together then this is really what we see is on a blockchain level account abstraction and if you then want to even go further and go outside of account of the blockchain level it's where you then also enter into the dep and offchain level where we say Obviously also deps nowadays assume certain things about accounts they use Easy recover to

recover your identity and check that you can log in in certain platform and that you have control over a key and also there there we really have to um make sure that we that this account abstraction comes into play where they don't assume that this is just a fixed key with a fixed curve but that we can say even depths should call into a smart account to verify okay how is this really a signature that belongs to this account and this is where you have the different levels of account abstraction coming in and most likely this will extend but yeah and currently most of the teams or a lot of discussion also especially in the recent day is actually about this blockchain level account obstruction how how can we progress here because this is really what will Empower smart accounts and really push us

forward to cover all the use cases that we had safe believe are necessary to bring in the next wave of users yeah and so we have this the core fundamental principle is about making this onchain experience from kind of these externally owned accounts these EA wallets to kind of this more smart contract based account wallets that that uses a different mechanism of of signatures and of um potential for automation gas tokens um and and and just broadly pushes that through exactly right like so I think the the prime example when it comes to this curve is always this when we talked about in the past at some

point Quantum Computing will be at the stage that ecdsa is not the right way to go anymore and currently that would mean we have to change the fundamentals of the ethereum blockchain and nobody knows how it would look like and they're having account abstraction where you say okay it's up to the smart accounts and smart accounts actually can support multiple signing mechanisms what we already see then this is a smoother transition right like we see this already that smart accounts that are out there um support mechanisms that use the sep1 curve which is the one that all the common smartphones support we see um mechanisms out say that use BLS signatures and even ZK knowledge proof right like and if you look at BLS and the knowledge proof that's where you

then go into the domain of being actually Quantum resistant and um we therefore would be ready for the future but at the core um eum has not moved away from that everything is based on ecdsa signatures so we still have this dependency okay give us like a quick like cryptography 101 when it comes to these curves because there's like these different curves you mentioned that there's an account it's made up of a public and a private but then to make interactions with those accounts we need to do signatures and these signatures exist on curves so how does this like how does the cryptography work here why are we choosing between

curves what's the difference why does it matter that we're on one what makes one curve Quantum resistant while another one is not I cannot go too much into details it's not the biggest domain but I very simplified speaking right like nearly all crypto graic like everything on a computer but especially in cryptography it's just you have different mathematic operations where you say in One Direction like you can do operations that partially you can revert to verify that they are correct for like and it's actually basic mathematic operation on very very large numbers and normally on very large prime numbers and to really do this back and forth calculation to verify it you you put them onto a curve and it's very important what curve you

use and what parts you keep secret and this is where then the first discussion comes in right like how big can these numbers be this is if you look at certain cryptog graphics and it's like they are 1,24 bits wide or 248 bits wi like The Wider normally the stronger your encryption because the bigger the numbers can be um and then the curve is really where you say this is when ecd is a you have a K1 and an R1 in the name of these curves somebody just defined random numbers that make up these curves uh the R1 curve is a random number nobody knows where it's coming from basically it's um recommended by nist which is one of these bigger institutions existing in the USA and where then people especially in the

crypto World say who knows where they got this number from might be compromised might be not fun discussions when you go in the especially in the etherum research um forums where this is being questions and there's a big discussion around this where the then shows basically Bitcoin shows the K1 curve where they said look this is how we derived it and therefore we believe that it's not compromised and then obviously these are quite actually not the young like all of this cryptography is not quite not the newest but then you have BLS which is a newer curve which um new crypto curves have always normally more advanced mechanisms for example for BLS it's easier that you can aggregate certain signatures which is if you transmit them over the Internet

obviously agregation is super awesome because you can better compact them um there are a couple other features also when it comes to how you can basically um use distributed key generation with BLS and so on there's a lot of research there but um obviously the newer it is the less you know about the curve right like if you have something and you see this also with safe if you have something that is running for years without any exploit it builds trust right every new cryptographic curve you always start questioning this again and then when we come to Quantum resistance it really depends how do you shange certain mathematic operations that make something Quantum resistant or not and this is where new cryptographic curves do different operations again here

disclaimer I'm not a cryptographic expert it's basically the oneon-one simplified and this is already the simplified version of cryptography but yeah and then I think the the fun part actually with smart accounts come into if you have something like zero knowledge proofs this gets a whole another level because you basically do cryptography in a secret environment uh you do it in a way that you can can prove that you did it but you actually don't share anything of what you did right like and this is there are fun articles here about where like what you can do but this goes a little bit beyond the scope here um but yeah so the importance always big numbers right like you want to have big numbers big prime numbers prime numbers are hard to find so the this is where a

lot of the magic are coming is coming from okay and and the fundamental change in in smart accounts and account abstraction are we changing the curve or are we just operating amongst this curve in a different way that allows us to generate these signatures more easily among different components so fundamentally um you can still use the same curves what you want to do is you want to completely be flexible what curve you use right like right now you say you use a specific curve like with a specific parameter um and what you want to do is just say you can use any authentication mechanisms that has a specific interface so in with if you take ERC 437 for example they have this um validate user

op um function and the contract then can decide do they use a curve do they just blind approve every operation it's also an authentication mechanisms if you want to take it like this do you use your knowledge proofs or do you have some fancy other way that nobody knows about yet right like but it's up to the to the smart contract and it's not dictated by the protocol and this flexibility um obviously then allows you to even combine multiple of these mechanisms and adapt new mechanisms once they come once they come up yeah yeah it's kind of a beauty of crypto is the optionality and it's the flexibility but often times that is the cause of the most uh friendly debate and research um which is which which kind of

brings me to a tweet that you put out um the other day with regards to um this idea of of the types of account abstraction and this kind of funny meme here um which was basically that there's devs are excited about 4337 but you know we've got this little pink guy here with 3074 um and I know this is comical to you but probably quite confusing to many so uh let's dive into 4337 let's dive into 3074 and it it was on one of our recent shows that a from haon spoke about this kind of dilemma as well so clearly it's

making its ways around the research Community um why don't we start with 4337 this was kind of the original ERC standard proposed um and now we're moving towards different solutions so um at a fundamental level what is 4337 and um maybe we can talk about your preferences as well yes so on a fundamental level 437 proposes some of these interfaces and standards that I mentioned before like so they actually quite to say here you have an interface um which def is defined when a smart contract implements it you can call into it and then you verify a signature and additionally to this they separate it quite clearly from the execution which is actually we are used to it from etherum right like um when you use a

normal smart account your signature is validated by the protocol then the fees are deducted and only then you execute um the the transaction data or the transfer right like and this is basically also what the protocol does it says okay we have have two parts here and one is validation and one is execution and and then additionally to this um ESC 437 also defines okay how do we allow smart contracts to submit transaction and they create basically this this alternative men pool which um has its own restrictions and where you also have your own gas fee Market basically and this is important because and this is also the reason why it's an

ERC and not an EIP especially makes this distinction very explicitly is that 437 is a way that we can collect information about okay how does it actually work what limitations do we have to put in place for this to become an EIP so that we can put it into the protocol because all this flexibility that brings obviously opens up certain attack vectors the eum is quite strict around like nones need to be monotonously increasing um because you can do certain shaks of up FR shakes up front that balances are not changing because you know your Esa balance can only change if your nons increases you can otherwise not lose either on and this obviously is

a deny our service protection so that we cannot just suddenly that blockchain comes to an hold with smart accounts all of this needs to be reevaluated and we need to collect this data it's a big research topic we there's a lot of things where we make assumptions but how it really works out we don't know and this is where 437 as ANC is super helpful it allows us to look into this and to understand it but to really get measurable data we need that projects adapted that we have enough traffic because if you it's like running a server and you say you know my setup's awesome and you have like 50 requests per seconds and if you go to any bigger team saying a 50 request you don't know if your server is really scalable or anything because it's no traffic right like so how do we get the traffic up and

this is um this is what we where we are pushing towards to like adoption right like and this is also where in the last year from um the safe team we said because you mentioned my preference right like why potentially for safe it's not the easiest to adopt 437 we have an existing contract it's a completely different execution flow it's hard for us to adopt and the immediate benefits are not um like not the Super biggest because you know it primarily solves for us the problem how do we submit transaction but for the wider account abstraction ecosystem right like this is necessary where we also say it doesn't help us if we say let's try to do our own thing and then as smart accounts as

a smart account equ system we do not move forward and this is really way say okay we need to engage also there we need to collect data we want to see how this pushes forward because our end goal is still that we have native account obstruction so this is really where 437 is like really the it's not the end goal by no means it's the starting line this is really where we start thinking about Native account abstraction it allows us to collect data and figure out how can we really solve it how can we push it n and you mentioned a couple of use cases uh that account abstraction enables um could you just kind of like go through what those use cases are that are not possible today but with account abstraction are now possible yes um I'm going to recount this a little bit

without 3074 in regards because this changes a little bit what is Possible only with smart accounts and what is possible with others but like in general with smart accounts you um um the one thing is that you can Define better the control over your I mean I mentioned before you have different curves but actually since you can it doesn't have to be only one key that control right like you can have multis you can have permissions you can have nested hierarchies this is what we're very used to from web 2 in organizations this is really where you say this is where smart accounts can put this on chain in a trustless way right like we know that this also possible with NPC to some extent but really smartcon Springs as completely trust

this on chain um another part which is um what you can also add there is actually that you have shs and this is where we really at safe say this is a pathway to um engage also in the wider ecosystem it's what we call hybrid custody basically because we all want self full self custody but self- custody is scary for a lot of people so giving them a helping hand but you say you have security shakes where where certain transaction need approval from more like from an external like insurance or something or like at least in the normal case else you need to have time delays and so on this is really checks that you can Implement and customize per user on a transaction

basis so this is there's a lot of flexibility in this regard besides this you can also do a lot of um convenience features that overlap with security for one is something like session keys right like also this something everybody uses in web 2 like we use o and basically every application where you can give permissions for a key that certain application can do certain actions um without having to ask username password for your whole account this is something where with um where with um smart accounts we can really say Okay certain actions we have like our special key and this can perform certain transaction with certain protocols without having to get the um confirmation every time like and another example for this is like

daily limits where you say um or recurring transfers like d limits you say up to a certain limit you do not you only need one Pro not multiple one and a mulc um re recurr transfer is also what we used to right like let's assume Netflix at some point would go fully on chain and you can go crypto payments we don't want to manually trigger this payment every time we would want that it automatically just gets executed once every month right um so these are some some of these convenience features and then another big section is recovery and there we really say um users lose their password they lose keys right like and this is one of the crypto things where we are all like we better not lose these keys

because else we're going to go to some garbage dump and try to find our old hard drive or whatever right like this is this gen General picture that everybody when we talk about crypto that's what they all think about what we're doing the whole day right like and this is where really also their um smart accounts can come in where you can say actually you can do key rotation you can um have mechanisms where you can say you trust a group of other people or another entity um with without giving them power over your account you can have time delays there are mechanisms that are called Dead Man switch where you say if your account if you can somehow prove that your account did not do any interaction for the last months then you're allowed to switch out the keys

like and there this provides just a lot of flexibility where we say yes self uh custody is awesome but it's also scary and if you really want want to get more people on board we need to be able to have the flexibility to really um provide the tools that it's less scary yeah I think the Automation and the social recovery were the two things that got me most excited but I think that the gas tokens um what is what gets a lot of people excited about not having to or being able to use any token for gas often times you just kind of get stuck in a chain where you don't have the token that this happens um and so we've see that there's a lot of these evolutionary benefits of smart accounts which uh it makes me feel like crypto is

still so early sometimes you don't feel like it's early but still feels early um and and yet there's still some better Evolutions to come from your perspective from a different model because ERC 4237 is a standard right and standards have to be adopted to be a standard if I if if I adopted and you adopted Richard then I mean we might as well just call it like a friendship right or a handshake so 3074 comes in um and you're excited about this what are what are some of the additional benefits uh which can be seen from from 3074 or kind of what are some of the pitfalls of

4337 which you find 3074 to be more appealing it's interesting I think maybe was not what I was yeah it's not what I was suggesting with my tweet um to be fair first first of all what is 3074 let's maybe start with this one like 3074 is actually um primarily giving away that us an EA can give control or authority that to execute transactions through your account to a Smart country that's a very interesting feature because in general if you talk about it in the smart account ecosystem this is to empower EAS it's not really empowering smart accounts because this is basically um allows you to do a lot

of the stuff like session Keys per what I mentioned before automatic transfers with EAS this is where we from Smart account say it doesn't benefit us like immediately the smart accounts but it's not that simple right like and um also the the meme that I was doing there it was not about this is 34 374 good or bad on a technical level because it's actually a necessary step that we have to do I would have done it at a different time but it's it's necessary step I think especially with Lucas together um we had a block post actually released today where we were talking about this where um the end goal is native account like we want that

on a protocol level it's not ANC but it's really on not protocol level we have everything is a smart account you have the full flexibility to do everything and there are many ways going there and the wave at least also in my head that we were going was okay we start with four3 7 we collect data with 437 then we look how can we push this NATA with rip 7560 for example this one native C of section on l2s if we have more data then we push it on Main net native and then we get to the point we need to provide an upgrade P for all the existing uas because they somehow need to become smart accounts too right um an other direction could have been that we say let's just keep empowering EAS until they become smart accounts it's also a

viable Direction it's and actually for us 3074 goes a little more into this direction and what then the challenges that we see is that okay if we go into this other direction we have to make sure that we just don't pull Focus away from the One Direction that this then slows down and we start working on something cool and new because you know it's immediate new thing that is there that we can go for 30 3074 definitely provides a lot of opportunities also since um it is more empowering uas big teams that are UA based um like metam mask for example will engage also in a lot more new user

flows that have been until now more exclusive to Smart accounts which will actually will create adoption also in depth what I what I mentioned before right like because with smart accounts it's also a replay protection can be different you can do batch transaction all of this until now was always for depth a hassle because they would have to support the normal flow for EAS that doesn't support this and the one for smart accounts and normally they say most of the users are still uas so let's stick with this and you don't get the full developer experience and this is really a benefit where you say okay we can go into this direction but especially at safey belief we would want that we also commit to the next step is um that comes after 3074 which is like let's allow a way to upgrade an UI

actually fully to um through a smart contract to smart account and this would be then EIP 5003 which also has its challenges um technical challenges there right and this is really where um this challenge where you say okay now thinking about this takes a little bit where the focus that actually the whole AA ecosystem had on 437 going down the road and this maybe not the best meme that I took there but this was basically my focus was not about is RC 3074 better than 437 or do they not like should do either or it's more like suddenly we had a lot of focus on 437 A lot of teams pushing forwards this and now part of them at

least will say oh let's ra look at this and let's focus on this one perhaps a a a meme Choice could have been the guy with the girlfriend looking back at the 3074 I think that one would have been a good one so it's not so much about the preference but it's more about like the timetable in which this process has to could take in that starting with eoas where all the users are you're kind of going where they are and also where all the infrastructure providers are and then trying to do that transition is going to be an easier process than trying to basically adopt a new standard then go into an EIP and thus the the paths could be a little bit Divergent if 3074 plays out how you see it leading into 5003 rather than going directly to

4337 yes and I think the the biggest challenge or the what we see as the okay it's going to be sound B but like the worst case what is that we actually stick only to 3074 and then as said 374 in itself does not provide many benefits to Smart accounts it actually empowers uas and what happens if you Empower uas is that people stick with EAS and it basically then in the end goal of getting native account abstraction then rather really create some inera that we then have to accelerate again right like and this is especially important to consider when we look at the hard fog road map of eum which is actually we have Petra which is the EIP 374 is now

planned to be included which is happening in the next six months more or less and then we have basically one to one and a half years which is allocated to work trees so that means after this next hard Fork there is for one and a half years nothing going in into further new eips that would go beyond 374 and this is really where the discussion comes okay can we take the next step right now too so that we do not get stuck for one and a half years um at something where we say okay now it's it actually makes it harder for Native account obct to really move forward and to to get adopted yeah and you guys are pushing pretty pretty forward on on um just

wanting to get the transition to Smart accounts as as as aggressively as possible I mean you guys are I mean this this this idea that you've now pushed out of key store rollups is like this this idea where it's you know I can control multiple accounts um you know and and execute simultaneous actions amongst a ton of other possibilities kind of from one Central Hub that that that that is very much so smart account oriented um so it seems like for you guys at safe um whatever is pushing forward smart accounts the fastest is kind of something that that you guys are going to be most interested in yes I mean kind of obvious since we our vision

is that every um crypto transaction is executed through a smart account even better so it would be safe right like so we believe really that for to get the wide adoption we require smart accounts we require this flexibility that is provided by smart accounts and this is why we are really pushing for this um especially when it that being said right like smart accounts are not solved we don't it's not that I could say oh they just use them they solve all the problems there are no challenges at all if this would be the case everybody would be using smart accounts there's actually a lot of challenge with smart accounts and and one of the challenges that we have seen spinning up starting mid of last year and especially now this year increasingly is that that we have more and more networks more and more

chains spinning up we have this cross chain experience and then if you go into this way actually with smart accounts there are two challenges here and or two key differences to UA since uh you are as I said very well defined what is the curve what is your public key it's also very well defined how is your address defined so actually your private key your EA has the same address on all chains this is not necessarily the case with smart accounts why it is possible possible to some extent often it depends in also how is the Smart account deployed what bite code does it have and then while there are mechanisms like create two which allow you to get on a lot of chains the same address actually

on chains like ZK zinc which have by default different bite code and because they run a different VM under the hood they not 100% evm there you always get a different address so this is the first problem and the other thing is obviously the EAS they are defined the crypto curve is defined everything is the same so they don't have state why smart accounts they have state so let's say simple example on mainnet um us three share small account and then we have the same account we actually manage on the same address same account on optimism to deploy the same but now we change on Main net you guys kick me out that doesn't happen automatically on

on optimism I I'm still part of this on optimism so now we have to go to optimism and kick me out there too and now let's imagine like the safe smart contracts they deployed on over 100 chains let's assume we are super active on all of these chains we have to do this on all of the chains this is not it sounds extrem this the key store roll up say so you guys have proposed okay we can't do this on the L1 though because it's a little bit expensive not all these users are going to be able to have a safe on the L1 to kind of control all of these different chains and all these different uh accounts so you guys propose what vitalic has kind of said as like this minimally minimally viable roll up or something to that extent

maybe talk us through this design so in general I think to start out with without thinking about if it's a minimal rollup what you want to think about is like you have one chain where you updated which is uh Lucas nicely put it in the word it's like basically your password manager it's like you have this one password manager and you have your one place where you update your owners for your smart account and then all the other chains just reference this one and this is simplest thinking about it let's assume it's actually it is on Main that you're safe and since most of the rollups actually can already have connection to L1 and can read State there they basically you just read the state from this account and use this for the verification when you do transaction that means as soon as you change your

owners on L1 it is immediately um like as soon as it's propagated to the L2 chains but then the owners are also changed on all the other chains without you having to go everywhere so while this is nice and actually since most of the chains can read L1 State this is actually decently good to do um also cost-wise the challenge is that it's just expensive to deploy on L1 right like so if you want and order to execute a owner change on L1 so this is where then V accs to really brings us to the masses we have to gas optimize this and the gas optimization here in this way would be let's do a dedicated rollup if you have a dedicated rollup we can optimize on really what operations you can do if there's really a limited set of op operations that you can do and

data that you can store it becomes more gas efficient etherum is also quite expensive because you can do it's a touring complete VM right like you can do everything which is for managing ident is not necess really necessary to do everything um Additionally you can structure proofs in a way that they are easy easier to verify right like um a z mainnet if you want to prove prove the stat it's a um Patricia Merkel tree which is actually not the easiest to perform a proof it's currently I think just to perform a storage proof so to check that somebody's an owner it's 300,000 gas just as a context right like a normal esc20 transfer with a smart

account is 100,000 gas or something like this so that's like you are tripling or like quadrupling nearly the gas cost so this is this really where then these optimizations come in but this is on a similar level as native account OBP is super early right like we first have to analyze how far can we optimize this rollup how can we optimize also the communication between L1 and L2 because even this right now is not it's non trivia and there are many teams working on this and this is actually what what is really nice to see currently in the ecosystem that even so there are the different teams besides safe we have B economy zero death um Alchemy all working on Accounts at the same time these teams come together and

work together okay how can we solve these problems because these are problems that are Universal to all smart accounts and this is really very nice momentum where we see okay this ecosystem the a ecosystem can actually push forward okay so let uh finalize the mental image of the architecture that that you just explained here cuz so basically we've got this key store rollup that is like a it is a rollup that does not have a ton of apps on it or a a whole lot of functionality outside of just being like this kind of password manager if you will this key store rollup is like this authentication machine it settles to ethereum L1 effectively uh ethereum L1 is able to um send out messages to all

these different l2s or rollups via like a state on like a block header or or like a a transaction hash and the block header um it telling those accounts to basically make any updates that you kind of request from your central password manager and then those accounts send a a state proof or a proof or storage proof down to the uh key store rollup saying like here's the updated State and then the key store rollup settles that back to ethereum and that's kind of like the general Loop did I did I mess that up I would I think in the second half it was a little bit I would formulate it differently so the first one is right

right it's a very simplif roll up that rolls up to L1 um and then actually ethereum doesn't send out messages it's more like that um if from any L2 you want to access the key store you basically read the state of your key store on L1 from your L2 so it's not that actually the messages are sent out you just have mechanisms that any L2 can read the state on L1 and this way you can basically and this reading of the key store is normally done why a proof whatever like in this case what V proposed it would be zero knowledge proof and this this one you can then take take a zero knowledge proof and you can proof on any L2 the state of your key store and here obviously you can do

certain optimizations how it's propagated down and I think one of the challenges is also that the communication how do you access on any L2 L1 state that this is not standardized and most l2s have this possibility because this is kind of required if you deposit funds into an L2 this is exactly what happens the L2 needs to know hey there suddenly my balance on L1 increased on a on this anchor contract so they read State on L1 and you can utilize this method to do the same than for the keyst gotcha okay so Switching gears a bit another fundamental part of this network is the safe token which is now transferable and so you guys aren't

going to use this safe token for any kind of Economic Security like maybe a cosmos chain that launches a token along with their Network well correct me if I'm wrong here but it it will be used primarily for governance and and social consensus and Social Security around this safe minimally viable uh key store rollup um I guess generally what is what is the role that the safe token will play in this in this key store rollup actually the safe to what role it plays in the key store up it's quite early to say there because the key store rollup is quite early right like where where we really went with the safe token is I mean the safe token in itself exist for

two years it has been non transferable and the big change is just that it gets transferable right like and what we see actually the safe token representing is really the safe ecosystem and there's a big safe ecosystem consisting of many teams and actually account abstraction is not just the smart account part that we talked about which is a lot like the key store roll up and this really the blockchain layer and also how you do certain um interactions with modules it's a lot of tooling that is necessary like we see it with 437 you have bundlers we see it with EAS even block explorers are necessary and we believe that to grow this ecosystem and in a sustainable way this is where you need some governance around like you need

some consensus and this is really where we see the safe token can come in and this is where um really with it becoming transferable this is where it becomes really feasible or like possible to really do this because to really enact this where you say like the safe token represented the ecosystem which is then also new people can join the ecosystem and the ecosystem can provide Security on how certain services are running on how certain modules which is like this extension mode that we use um for smart accounts like how are they secure or not to give guarantees this is where the token needs to be transferable and this is really now that we flip the switch it's actually now we can really start saying okay let's Implement all of these

things and we have been working for the last year on all on many fronts right like and actually it was not the key store rollup which where a lot of this has been talked about it was more like this security of the modular smart accounts which I think half a year now ago that we proposed or like that we published this safe protocol paper and this safe protocol was like this high level Overview at say okay how do we see modulized smart accounts because it's similar with 3074 if you give control towards another smart account that's a big security risk like because you have to trust this other smart account and in 3074 there basically it defines it yours wallet for example metamask trust them they will

tell you what are good um what are good smart accounts that you should trust but we inherently believe that it cannot be a centralized party that should tell you what is a good or bad smart account that you should trust and this is also where an ecosystem comes from where you say look we need to govern this we need to provide security we need to provide guarantees and this is really where the safe token then will play a role to provide these Securities in the key store rollup actually we are so early that I could not immediately say where it plugs in there but also there um the challenge with the key store roll up as with many will be economic incentivization this is it's not for free but in the best case we would love that it is for free to the users and

also there the token can play a role where you say you know certain parties earn from this and you want to redistribute it so that you can certain parts of them of your ecosystem can run for free um and this also this is something that is not specific to safe and like the key store rollup is um our discussions that many teams are having for different parts um also in regards to 437 um and transaction sponsoring right it's one of the Prime examples for 437 I remember the token coming out um yeah like you said year and a half ago two years ago maybe in the bare market and like it was there and and I think the the token amount per safe was like 2,500

tokens or something um maybe I'm mistaken I don't know could you talk us a bit could you talk us through a bit about the distribution how uh NOS safe owners um you know kind of got tokens and um kind of what that process has been yes um it was nearly exactly two years ago uh next week it will be exactly two years ago when the token was launched um and there were let's say um I think four four bigger pools five bigger PS so a big part actually went to the Dow it's like 60% of the tokens we had a part for the team and for investors that is allocated um then we had a big part also for gnosis since we are we're spinning out from them so

gnosis St also also a big part of the tokens and then we had also big part for users and ecosystem and think this is where I want to focus on because this is the one where really say let's get them involved when it comes to the users um we actually only there's still quite a lot of tokens allocated to Z that are still available for future air drops um because the first part was primarily focused on um e main net right like we wanted to drop to e main net because this is where most of the traffic when safe really started was happening and that really we said okay how long does the safe exist how many funds are you storing and how many transaction are you doing these are like

there were some analytics done and this is where really you actually there was not one fixed value per save but it this really depended on how much safe you get because for us we believe in that it's just having a save is not enough right like you should store funds in them and in the best case you should use the funds just holding funds is not in sense this is where we had like this very long list it's actually public in our repositories for the airdrop where you can look it up on all these air drops that were happening and then the second part was an ecosystem air drop where really said okay let's talk to like actually everybody could apply and and say whoever contributed to the success of safe you could say I did some educational videos I did some tutorials I built a project on top of you or I

actually build a full running production ready product on top of you right like all of this um like was giving out a grant and this was actually quite a lot of save tokens um I think the highest one was six digits somewhere safe token wise right like you can look this up it's also public information um I think the the biggest contributions are like one 1 million save tokens plus some something nexist so this is really also where we said okay we really want to continue building with these people so we give them safe tokens that so it's soon as we really get into governance and as soon as there's um really tokenomics that these people can participate in our ecosystem and now with it being transferable this is really the point

where we can start doing a lot of more Dynamics because as mentioned for example the user air drop was main that only partially also not exclusively but partially also because briding the token over or launching like another token which is also safe token on other networks is a huge overhead which which provides quite some distraction but now that the token becomes transferable it actually can engage with ecosystem and say let's bridge over a certain pool of our token to a specific Network and do an air drop on this network for the people on this netwk like so all of these possibilities really engage us to allow us to engage with our ecosystem and really get more people involved and this will create the power in our opinion

gotcha and are the are the tokens claimable uh for a um a set amount of time or is it in perpetuity like for for those who are watching now um obviously the token's just gone live in terms of transferability today you know how long do people have to check is it too late you know people are going to be wondering how they can check their out allocations and kind of you know see what what they have so the air drop actually happened two years ago right like so this one is actually not available anymore will not be Shand are that there will be there were actually two air drops so far and um I would have to check the second one but both of them

actually they're okay I have to explain our air drops so our air drops as you get them allocated you then have to say um like half of the tokens you can immediately claim and the second half would vest over a couple more years but to be really able to eligible to claim the vested tokens you have to say yes I would like it's an active opin it's not forever there because we say tokens that we allocate to an airdrop and that just sit there forever to wait to be claimed that is not feasible we rather had after a certain time we can the tokens that have not been actively claimed we can Channel back to the ecosystem and do another air drop for active people right like so actually the two air drops that we did both are over this period and currently you cannot claim more air

tropers that being said also very um like like the the like in our governance process right now now the we had this part where we voted on transferability and now we like since it was today so at the day of the recording where we the the vote was finalized and now we have like a 8 Days period where it cools down before it really went goes into effect this is a little bit related to our governance process because we actually do off Shain voting with snapshot and then we have to use an like an oracle which uses an escalation game to get the result on chain because again this is a fully decentralized process it does not rely on a centralized party being the

trust holder but it's really where we say we use reality IO um reality. e where you can then play the escalation game and have onchain and Trigger the execution without any involvement from either the founders or the safe ecosystem Foundation yeah it's that's really the way to get the most decentralization out of the governance process um I I appreciate you spending the time with us today uh it's been it's been a pleasure to learn about the safe ecosystem as well as uh the differences in in account abstraction I want to close the loop on the 3 3074 conversation and and the 4337 conversation because we brought up earlier how these paths May diverge I'm

curious do do you see these paths diverging over time or do these paths ultimately converge to Smart account and Native account exraction ultimately I believe that they converg again like it's actually the diverging is more the short-term effect right like and this a the challenge for me is that the short-term effect is what happens now and this is where it's like um where there's a challenge and where we will see impact which might slow down certain processes and will change certain dyamics But ultimately I believe that like both ways go through native eon abstraction if we continue on these paths like if we don't stop halfway

through um the thing is that both path no matter which one you take if you take 3074 and then go for um 003 and continue on further or if you take 437 with rip 7560 and so on they are hard ways it's by no means easy this is really where we would there's a lot of hurdles a lot of things that we don't know yet and this will require some endurance that we go through and this is this is what we need to do this is what we need to commit to that we continue walking and that we don't stop halfway through and then we have this incomplete account abstraction because this is where then if we stick with um RC 437 how it is right now then this will not have a long-term effect

people this doesn't work on Main net it's no secret it's very it's just too gas expensive too much gas overhead to work on Main net really properly um and it's not meant to be if you talk to UF he will tell you and like when we talk together we all agree it's a stepping stone to proper native account of striction but this means we have to go the way and this requires Manpower and this requires also that we push together and not try to be all over the place but yeah we all have the same goal right like we want to go towards Native G obstruction and both ways will Le ultimately awesome yeah I I tend to agree I think ultimately we're stronger together and it's important not to split our focus and try to take both paths in parallel but that a lot of times that's

the nature of experimentation is that we we'll kind of run both of these experiments in parallel and the one that seems to be a little bit more promising will probably absorb a lot more of the Manpower and uh we we'll start to see a lot more support around that maybe right now that's 3074 before it was 4337 maybe it'll be 4337 again but ultimately both these paths lead to Native account abstraction and smart accounts yes I mean it's also so there always the Dynamics right like new stuff it's like a new toy for kit it's first excitement and then after a while you realize H you know this toy is it's actually not that exciting and then after a while you feel like ah it's actually you know I can do a lot of stuff with it it's like this curve like

Lucas found this very nice curve here excitement hype and then it goes down and then it goes up again and 437 as you said it had hype last year and it got into the sobering face right now where everybody's like is it it's not really that easy and currently 374 has this yes it's hype right like it's definitely hype and we will see how it turns out but there are chances that it will also go there um to the more sobering phase um that being said right like I I have to admit right I maybe I'm too pessimistic coming from a tech background and also being like running a business to some extent I feel Focus as you said is important that being said eum ecosystem is Big right like there

are a lot of developers it's we can do certain things in parallel it's more this my Thrive for efficiency that I would love that we all work together on the same thing but you know it's it's not that black and white yeah thanks Richard for coming on and sharing some insights we to had some really key takeaways here and I'm uh excited for the future of safe and um great work so far I don't know how you sleep at night uh the entire defi ecosystem is literally on your and Lucas's shoulders so uh I we we commend you for your work sir thanks was a pleasure thanks for having me thanks Richard