Ethereum is for Argentina - Vitalik Buterin - ETHCon Argentina 2025

Transcript

Okay, thanks for the intro. Welcome proof of stake.

Yeah. Something I always wonder about like uh you speak multiple language languages. Um, did that have any influence or relate somehow to the fact that you build a protocol with a decentralized nature? Like you're a person who likes to travel to visit different countries. Is there a relationship between the two or is just correlation without causation? >> I'm sure they're all connected to each other. >> That's good. >> Okay. Well, let's go back 10 years when you first created Ethereum. Uh, I'm curious about what was missing from Bitcoin that made you create Ethereum.

And I'm not looking for programmability as an answer. I'm looking for the specific thing that you wanted to build that Bitcoin didn't let you. >> Yeah, I think I mean one of the specific things that I remember being uh excited about was DAO, right? Decentralized autonomous organizations. This was this uh idea that uh you could basically take the uh the logic like the rules of how something like a company or some other organization works and uh you can put those rules on the blockchain and you could have uh you know some kind of company or some other kind of organization like actually run entirely on chain and uh I remember I mean I was excited about many different things

right like if you read the uh Ethereum white paper you know there's like 10 or 15 different examples but if I had to Epic one. I think that was it. And uh I uh remember I mean like back then there was uh there were like protocols like bit shares that was uh trying I mean back then Daniel um you know he used DAC's decentralized autonomous corporations and uh he was basically saying like let's make a separate blockchain for each DAC and uh I had this idea that like actually they could be more secure and like more trusted and and interoperate better if like we actually had one blockchain that could support all of these things as programs

on top of uh on top of them. So yeah, and thank you like if if there was one single thing, it would be that >> actually that's a good segue into the next thing I wanted to ask you. Like >> if this was the use case you were targeting, why build a general programmable platform as opposed to something specific for this use case? what prompted you go to the extra mile and build a blockchain where you could run anything? >> So, right before I started Ethereum, I was working at uh um covered coins and Mastercoin. There were some of these protocols that were starting to do these like more complicated applications on top of uh Ethereum. But then I noticed a problem which is that uh people would

constantly come up with some new thing that they wanted to do. And I saw that uh like if you did that on top of these protocols like these protocols would have to change and add a new feature every time someone wants to do some kind of new thing. And so I realized that like if you like if you wants to make it truly permissionless for people to invent and to do new things, you would have to uh like actually make something that's general purpose that like has this programmability. So that if you do come up with something new or and like even like within uh DAO's there's like so many different ways to uh to construct one. There's so many uh

different governance mechanisms. you can have so many different things that they can do that uh like if you don't want to have to keep changing the blockchain every week, you know, like you just have to make a general purpose. >> And I guess you're absolutely right on that one. Um back then when you launched first launched Ethereum 10 years ago, like what did you think that was going to happen 10 years later? >> Like what was in your mind that would happen at the point in time where we're now and how much have we drifted from that?

I really did not know. I mean, I remember when I first started Ethereum, like just writing the white paper at the beginning, like I really expected this would be a fairly small project, like we would finish it within a few months and then go back to university. And uh and so I think we really uh underestimated, you know, both the complexity of actually building something like this and how much impact it would eventually have. >> Makes sense. And I'm curious like if the not talking about the technical side of things but more the cultural side of things.

>> Do you feel like the spirit, the culture, the ethos uh that makes Ethereum what it is today was the same back then? Did it change during this 10 years? Have you changed your mind about how you see Ethereum? M I mean I think obviously the uh culture of the crypto space has uh changed a huge amount over the last 10 years in I mean all kinds of directions right and like some of them I like some of them I'm also critical of and uh I mean the biggest change I think was like 10 years ago people did not really I mean one is like people did not really know exactly what they're building and uh the technology was not really ready and so we saw a lot of experimentation a lot of

prototyping a lot of things backed by like future promises and future potential but uh not really uh something that could uh actually um like have a big real world impact at scale. I mean obviously except for you know the store of value use case that one was the most successful the first and now technology is much more mature people are doing a lot more things but then I also sometimes uh like miss the spirit of exploration that the space had earlier on like I think there's definitely a big part of the space that's kind of gone too much in the direction that uh you know we understand DeFi we understand

memecoins, we understand uh 50x leverage trading and uh we uh even understand ENS and uh there's like you b like basically if you wants to succeed you just uh like be a fast follower and like copy a category that's already proven and uh like I see why you would wants to do that from a like just risk aversion perspective but I think um you know it also uh does kind of harm the imagination of uh of the space and uh like I think it would be nice if we also went back to you know like people trying to create decentralized Uber uh people trying to create uh totally different kinds of companies that engage

contributors in totally different ways. Um and I think some of the new technologies that uh we're including like especially things like ZK you know they can help uh a lot because uh like they're actually it's introducing a whole new dimension of capability and with that I think there's a whole new dimension of uh things that uh like you can explore and things that you can uh like actually make happen right like uh if we look at even things like Dows right uh you know in countries even in companies like everywhere like decision- making is not fully transparent and sometimes that's a problem right but sometimes there's very

good reasons for this like you do not want every single person's vote to be public right and this is something that uh you know blockchains were not really capable of uh doing and so we saw these dows right and like on these dows all of the votes are public and then like a lot of people complain about Dows right and I think especially about a few like at the beginning of the year there was this like peak of disillusionment where a lot of people gave up on Dows and uh there were many complaints but one of the complaints is like every vote is transparent and so the whole thing just turns into like a big social game and like everyone is uh you know like follows along you know like the powerful and cool people and everyone's afraid of going against them and uh it just

becomes not very interesting and if you add privacy back into voting which like projects like Aragon for example are actually know doing in other projects as well then uh like you can actually fix some of those problems right and so when there's this new wave of technology and even when uh existing technologies are improving like how ethereum is uh scaling uh quite a bit and uh it's uh you know scaling I mean both you know L1 and L2 and like the nice thing about the L2s right is that I mean even if you know you personally don't like any of the L2s the fact that more activity is happening there actually means that more spaces for Yan L1 right and it's actually like extremely cheap to send

you know Ethereum transactions now right and then scalability will go up another like 10x um I think pretty soon and so that also opens up new possibilities for what to build and so and then also you know user experience and security and like web 3 web two integration and so a lot of uh ideas that people kind tried and then gave up on like I think uh it makes sense to really yeah try to bring many of them back >> and I think you're speaking to the right crowd here with a lot of new of enthusiastic people who want to jump into the space want to go back to privacy for a second um I've been working on that for a couple years so it's something I'm interested in like you mentioned that early on like privacy

was not built in in many of the project many of the platforms do you think it was just a technical limitation or was there something else >> I think technical limit ation. I think uh had zero knowledge proofs existed at like the current level of quality, I think Ethereum would have had them from the beginning. Um like but you do have to remember right like the technology was uh like what I remember was uh 6 months before Ethereum started there was this big uh Bitcoin conference in San Jose, right? Um and uh at that like this was the first Bitcoin conference of like any significant size, right? And I remember going there. I remember being really

inspired. And I remember the conference really convinced me that like the space is actually real. That it's not just people talking to each other on the internet. That there's like many hundreds of people that are have joined big Bitcoin companies that people lots of people pouring their lives into it. Lots of people making real startups and real products. And uh I also remember there were the people from Starkware I like Ellie Benson and a few others and they were making a presentation on uh the the some of the very first zero knowledge proof protocols right so there was actually there were two I remember right there was uh the the work on uh zero cache right this was uh the I mean

Zcash of course I mean like is what uses a later version or used a later version of the zero cache protocol, right? And then uh since then they replaced the entire technology stack with something much better. >> Mhm. >> Had zero cache was like the very first you know theoretically usable protocol for these zero knowledge proofs. And the second was a skip, right? This was what Starks were called before they were Starks. Succinct computational integrity and privacy. And with even with Zcash, so like three years after that, yeah, >> it took two two minutes to to sign a transaction on a big laptop and on a phone it was impossible, right?

>> And so if you fast forward to today, today you need uh like on a laptop like less than 1 second uh to generate a proof and uh I mean and then on a phone it's like two or 3 seconds, right? And like and even those numbers, you know, they're going down very quickly, right? And so and then also for a developer experience right like before 2020 if you want to make your own ZK stark application like you have to personally be a cryptographer you have to personally expand understand the details of uh all of these uh different spaces and like how elliptic curves work and then we have circom and then we also have noir we have many things now and so for a developer the yeah experience is

much easier >> like I know that you're pushing for a lot of efforts for adding privacy to to Ethereum. But >> if you had to redesign Ethereum from scratch, how would that look like? >> Well, in terms of privacy, I'm asking. >> Yeah. I mean, I feel like I have, you know, like a bunch of different answers and of, you know, to how I would redesign Ethereum in general and like actually I think the lean Ethereum road map that exists right now like already >> incorporates a lot of them. the one one of the big questions is uh you know should Ethereum uh just uh like should Ethereum actually have like explicit privacy features on the L1 or should Ethereum just do programmability and

then like basically be sort of fully application agnostic and you know I see arguments for both of these right I think uh the because even if Ethereum does not have privacy as like a protocol feature you could still have a lot of privacy on Ethereum, right? Like two days ago, we announced, you know, Kohaku and like you can have wallets with privacy built in and then if you use Kohaku, you know, like it like you basically get a uh like an experience of sending private transactions that's uh like not like fundamentally not different from you know like what you would get in something like Zcash, right? And but uh but then the like the benefit of uh like putting like that kind of support

directly into the protocol is like you get much larger anonymity sets, right? So like what an anonymity set is it's like if you have some kind of privacy protocol and like this is not even just about coins, right? This even applies to like VPNs, you know, like privacy on the internet, anything. but then only a thousand other people are using that protocol. Then if someone sees that some message was sent using that protocol, then they already know that there's only a thousand people that could have sent it, right? And then you can only add in like you just add in a few other bits of information like, oh, these three messages were sent by the same person.

And then you look at the time zone and like these people were awake. You look at the language and then like you narrow it down. you narrow down and it's like it's very easy to determine exactly who did uh any particular thing, right? And so the only way around that attack is to have like a larger base of users, right? And so if uh you know like Ethereum had much more privacy like baked in as a protocol feature then basically the number of users by default would be like every Ethereum user and but then the downside is basically that uh you your security becomes much more dependent on the technology in like a really scary way right because uh if you if there is

a bug in the zk snark mechanism Then basically anyone has the ability to secretly print an unlimited amount of ecos inflation back. >> Exactly. They can basically hyperinflate Ethereum. And the problem is like you can't even revert because no one knows when the attack started. And so but if um you know you have these like second level protocols like if there is a bug in rail gun then only rail gun users will lose money, right? And so it's like much more contained. And so to me, it's like a question of technological maturity, right? Like the more mature technology gets, the more okay we are with trusting it for more and more

things. And so I think over time like it will make sense to make Ethereum a default in more and more places. like eventually I think uh like privacy at all one kind of EIP makes sense and uh I mean even like things like uh you know like our effort around wallets is like a big intermediate step right because we're going from saying privacy requires a dedicated wallet to saying privacy is default at the wallet level and then the step after that which could take years but that's fine is uh you know privacy being a default at the protocol level right and so yeah I think it's like a matter of just being responsible and like understanding, you know, like which stage you're at in terms of uh risk and like being more and more comfortable

over time. >> I have so many follow-up questions. Um you mentioned maturity and well oification has been a word that has been thrown around especially in Ethereum in the later stages. >> My question is how do you feel how confident do you feel aroundation given that we don't know what we will know at that point in time? Like right now we're saying okay with everything we have learned these past 10 years we think that making this set of changes and then osifying will be good. >> Mhm.

>> But how do you know that you won't learn new things in the next three four years that will make you keep pushing oification back? >> Yeah it's a it's a good question. I think uh the like oification is also not a binary thing right and uh like we can be uh more like oifying the consensus layer and uh oifying the EVM or different things like you know you can oify a virtual machine before you oify the consensus. I think uh the uh mentality that uh like I've had is that like I think more and more oification over time is good for Ethereum. Okay. And I uh I

do think uh that uh you know we have like we we have a much lower like rate of surprises now than we did you know like before right like at the start of Ethereum we did not know any like I remember the blog post I wrote on proof of stake in January 2014 and in that blog post like I was hedging because I had no idea if proof of stake could even be made secure right and so then five years later you know we knew knew that theoretically ZK would change a lot of things, but then we had no idea when ZK would come, right? And now we know kind of we know a lot about what ZK looks like and even like what kinds like how efficient it is and how efficient it is in different situations.

>> Um maybe the one thing we don't know much about is obiscation. Um but uh and so you know like it's still good to maintain some flexibility but and then on the consentus algorithm side you know we know what kind of optimal two round protocols look like where we know we know what optimal one round protocols look like the difference between like the theoretical like optimum that we like that we know that it's not possible to be faster than and the thing that we know how to build as like less than a factor of five probably. Um and so yeah like it's it's it's one of these uh you know like things where sort of you don't want to like do it uh too early but at

the same time like as Ethereum gets uh more and more mature you know the problem is if the uh if the protocol keeps changing quickly right then you know like that itself you know like it concentrates political authority like it uh it it concentrates power in this governance body that you know keeps on deciding some pretty significant ificant changes. Um, and so that's also not a thing that you want. So I'm, you know, like I'm in favor of like I mean basically you know the road map we described yesterday, right? Where like we do the short-term scaling things that we already know. We do the lean road map that we already know the like the p the things but we don't necessarily know every detail in some cases. But we also

know like if nothing got if nothing new got discovered like what like what we would stop at. Um and then after that point like I like I think it's uh it would be healthy to start moving attention off of the L1 and start moving attention to some of the more surrounding infrastructure which like we have been doing already right. Yeah. So like I think like I think continuing that trend is a healthy thing. >> Surrounding for you mean swarm whisper >> swarm and whisper is one example. um you know and you know Exur knowledge proof related things um user level stuff like making more secure wallets, making more secure light clients, uh privacy infrastructure

um like in five years uh like if we think about like if we think about it in this kind of very like pragmatic perspective of like the Ethereum Foundation has a budget of this number of dollars. This is where the Ethereum L1 is now. This is what what we could accomplish by making Ethereum ML1 even better versus this is what we could accomplish by making the best possible infrastructure on the user level, maximizing user privacy, possibly even putting some into like more secure operating systems. Um like then uh like there will come a point right where the lowhanging fruit are not over here. The low hanging fruit are over here. And uh yeah, so to me that's uh kind of the big

uh like reason why like yeah I think short term this uh attention on doing you know like this big set of uh changes to L1 is good. Uh but then uh like o over time you know like the application layer is where more and more attention needs to shift to and like and you know like that's where this year things a lot of things have already started. sense you have touched on a lot of different parts of the of the stack. Um friend of mine told me the other day that um in mathematics it said that uh Oiler was the last mathematician that actually managed to get a grasp on everything that was happening.

>> Maybe because he was really prolific. Don't know but math hadn't advanced that much so that a single person could still cover all ground. >> How do you feel about crypto? Like crypto moves at incredibly fast. Do you still you can still keep a grasp on everything that's happening or are you starting to have some some blanks around certain areas? It's definitely more difficult than before but like actually not by much right because uh this is like the thing that I've noticed is like as these these algorithms progress I think uh sometimes they get more complicated but also sometimes they get simpler right like sometimes the complexity is just a discovery process right and then

sometimes like optimization is simplification like you give one example right like elliptic curves and especially elliptic curve pairings. >> Elliptic curve pairings are like crazy complicated. Like elliptic curve pairings are the one thing where even though I personally wrote an article about how they work and even though I personally implemented them, I have no idea how they work and >> now I feel less bad. Okay. >> Yeah. But uh the but then what's happening more recently is like I think a big part of it is like qu because of quantum resistance and Scott Aronson uh this like longtime quantum researcher wrote an article where now I think he

literally he literally said he thinks there's a substantial probability that we'll have quantum computers running shores algorithm uh which can you know like break elliptic curves in RSA before the next US presidential election which is like a scary timeline if that's true. Uh and so elliptic curves are going to die, right? That's like I mean obviously quantum computers are unknown like still they could come in four years, they could come in 40 years, they could come in 400 years, but uh like the things that I've lately been hearing are like you know it's starting to be more on uh the uh uh I mean I don't want to say you like use the label as optimistic and pessimistic because these are like very double-edged swords, right? It's like uh

in it's I mean not as bad as nuclear but it's like nuclear you know you get the power plants and the weapons and like here you know with with quantum like you get you you know you get these more efficient search algorithms poss but on the other hand like they just like break like sweep away about like half of all the cryptography that we use today right and uh the good news is that we can adapt to it right uh but the bad like it is a lot of work and then also the newer things are sometimes they are less efficient like lises are less efficient than elliptic curves though sometimes it's more efficient like uh people are moving to from snarks to starks even today right because uh

starks they're actually much faster to generate than uh elliptic curve proofs and uh yeah so like basically you know like as a result of all of these shifts it looks like the the amount of mathematical backgrounds that you need to understand all of cryptography. it might actually reduce which is uh you know really cool and like it makes me happy right I think to me a bit like a dimension of Ethereum's decentralization that people forget often is like the number of people who can understand either any individual piece of the protocol or even the number of people who can understand the whole protocol right and uh you know there is kind of the the sort of pragmatic mindset that

like oh that doesn't matter like as long as different pieces are are understood by different people like the like this is how society progresses and it gets more complex but then I like I like I can see that but then I also like I really you know feel and like see the value of kind of Robert Heinline's approach where he said like he famously said specialization is for insects so I uh you know I think yeah people being able to understand like all of Ethereum on a technical level is like part of Ethereum's decentralization too and like lean Ethereum really cutting down the assumptions and like making everything based around the same type of

hash as like actually something that excites me in that way. >> That's good to know. Yeah. And it's very very clear and very apparent on the lineum road map that that's the direction. >> I feel like I need another hour just to ask you about onchain democratic elections and simplification who can audit but we don't have that time. So, I want to I want to shift gears with the last minute we have. >> Um, I had so many more questions I wanted to ask, but never mind. >> Um, it's an excellent question. I always say it in English because I don't honestly don't know how to say it in Spanish. All most technical terms we say we say in English. Uh,

I'm just kidding. >> Okay. >> Um, no, honestly, I don't know. And I I'll look it up, promise. Um, last 10 seconds for all the people new here uh to the ecosystem. Any tips on how to make the most out of this week? Um I think uh I mean it all depends on like what uh your kind of position is and like what your personal skills are. But uh I would say uh like try to like ask yourself the question actually this uh like I remember this was someone gave me this prompt and it just like inspired me to really get into Bitcoin like don't like ask yourself the question not just uh you know like what

have you learned this week but also like what is one way that you can actually become part of this community and part and uh can and actively contribute to this ecosystem. I yeah remember back in 2011 uh the way that I started was uh I uh started uh going through the Bitcoin forums and I found a a Bitcoin job. I found uh someone was writing a magazine called Bitcoin Weekly and there was paying Bitcoin for people to write articles for the uh magazine and it was like a very small start but from there I learned more and then eventually I got into programming. I understood Bitcoin more deeply and uh you know eventually did Ethereum two and a half years later, right? And then so I think for anyone

here like think about that question of like what is uh you know like something that you can uh do to become uh a much more active part of uh the ecosystem and like listen to you know the presentations and like talk to people and see the uh exhibitions with that in mind and I think um like many of you will uh come out with a very good answer. >> Absolutely. Thank you. And uh I you were also mentioning about all the extra infrastructure being built about about Ethereum around Ethereum all the new progresses and we're still early on those like there is still plenty of room for innovation. I think that people new to the ecosystem can thrive there and push things forward.

>> Yeah. Perfect. Thank you. >> Awesome. Thank you so much Vitalik.