Ten Years In: Ethereum’s Next Frontier - TOKEN2049 Singapore 2025

Uh a lot of what I want to talk about is not only just reflecting on the last 10 years of Ethereum, but also kind of how do we think about the Ethereum ecosystem moving forward. And uh I don't want to spend too much time on the past, but I I actually do want to spend a little bit uh of time kind of going back to uh the early days. And I want to start with you Joe by asking kind of 10 years in, what do you think has Ethereum achieved beyond your expectations?

So my expectations were pretty high. Um, but I could take a stab at it. Um, so back then when Ethereum started, uh, Bitcoin was was quite young and it was, uh, um, it was mostly crypto anarchists. Um and um the ethos was kill the banks uh shut down the central banks. Um and you know decentralization uh needed to start that way.

I think um uh Ethereum uh had its share of of crypto crypto anarchists, but I think the they whoever they were were were more pragmatic. Um

those who shall not be named.

Yeah. Uh someone who is not me. Um and uh uh and um and so back then I think there wasn't really the belief amongst any but a very tight group um that decentralization was a viable thing that could um be merged into the real world at at any point.

Yeah. My my expectations were set really by Joe by Vitali by all the visionaries. uh the expectations were really high but I think that also we were observing the world changing uh we observing all of those things happening and yeah I felt that the institutions will go through that evolutionary phases of getting the adoption and and I felt that also I felt that it was my mission of moving from the institutional world to to start building from zero by by looking at the white papers yellow papers and building from zero through my my own implementations and feeling that it's really permissionless and I'm changing my own uh old world into the new one and I'll observe everyone getting on board gradually and my mission to make it evolution rather than the revolution because as Joe you say you would expect the cipher bank to kill the banks and and that sounded like revolution right but Bitcoin was much of this at the beginning and I felt it doesn't have to be like that and uh you know many of the crypto people refer to to the books like foundation and a foundation was very often talking about how to shorter that shorten that barbaric times uh remove the revolution and just let lead the humanity to uh to that progress to to shorten the times of uncertainty and I I think that's what we are doing by bringing everyone together bridging it uh Ethereum is fantastic at this of uh welcoming both the cipher bank people and the large banks and the payment systems and all the users so it's great

might as well I'll let you jump in too Yeah. Oh yeah. Um I came into uh blockchains 2018 and when I looked at the Ethereum road map uh and you know the one of the main things there was zero knowledge proofs zero zk rollups for scaling Ethereum you know even for an academic like me it sounded too far-fetched that you would take an idea like that which is very much in its infancy and then say hey I'm going to bet this platform scaling on a technology that's never been built before. And today we sit here you know Lena as a ZK roll up. We have so many amazing teams actually built you know all kinds of zero knowledge scaling technology where you can do complex computations and then submit a simple proof the computation was done correctly and you know it's to see that live and actually like change the world as you know the question asked is insane.

So that's what has widely exceeded my expectation. I mean speaking of I guess like maybe I'm kind of curious actually for Tomas like is there anything else that uh that you feel like is no longer true that was true 10 years ago about kind of how we thought about computing in and crypto?

Yeah, I think we changed the world. So we 10 years ago we said uh the largest threat is a centralization of financial systems and now we feel like we we already see the path forward which is no longer centralized financial systems. It starts to feel obvious. it starts to feel like we know the path is just integrations bringing everyone on board but we know how it happens. Uh it means that that world doesn't exist anymore but we changed it.

It's not that the the world was changing and now we just lost because we thought it will be different. We made it we made it happen. Everyone here was participating in it. Uh and I think the next 10 years hopefully will be the same in a positive sense that we think there are some threats there is AI centralization. There are big challenges for the humanity and through this technology we'll participate in that change again and in 10 years we'll say that that world is that world is no longer there but that vision is here with us.

That's amazing. I'll do the honors for you Joe which is what do you think is no longer true?

Sure. Um so in 2014 um there wasn't the sense that everything was breaking down. Um so in 1980 uh life was you know gentle, warm and fuzzy in the '9s you know uh we gradually as uh the 80year super cycle from from the end of World War II wore on. um we got into um more and more edge cases in terms of how the system of the world and technology was serving humanity. Um by the time the Patriot Act was passed in the United States, uh things started to get a a little wonky.

Um top down command and control started to get a little too controlling, a little too capable of of controlling. And so um you know when financial crisis started to develop um in the as um because there's just too much debt in the system, Satoshi led out a cry uh in the form of the Bitcoin white paper. U but still we we hadn't yet experienced COVID. We hadn't been locked down. we hadn't had cities, businesses, industries shut down, people um losing their businesses and their jobs.

So, it was still a a much more innocent time. So, there really wasn't the need for this solution. Um and now there's a desperate need uh for us to evolve society, the global economy uh in a more healthy direction. So top down command and control is really great for building civilizations. It's really shitty for 99% of the people in those civilizations.

And so uh Satoshi invented decentralized trust, showed us how to operationalize it. Vitalic and many others um showed us how to build um with the technology and so many different functionalities uh with decentralized protocols. uh and that that adds a a horizontal organizing capability, a horizontal layer to the vertical layer which works really well. Um and so we I think we have ways now uh where we can work our way out of this mess and um Tradfi and the mainstream is starting to to really understand that, appreciate it and ask for it.

It's been a very uphill battle, but now you see acceptance and that's kind of interesting.

Yeah.

Um okay, enough of the past. Let's uh let's jump right into uh now and and the future. Um I'll start with Tomas. So scaling blobs, UX, and scaling the mainet. These are Ethern Foundation's three top priorities.

Um my question is how will we know that you're on track? And uh and if if there was a report card, what will we get as grades right now?

Yeah. Yeah. If if you were getting a report

uh we we published in uh August I think uh or September like August September we published three blog post for each of those tracks. Okay.

So each of those tracks have their respective two track leaders that are on the edge of research and engineering. Uh and they publish the progress and uh they keep sending reports uh to everyone on uh how how much L1 is scaling. uh we have now the tracker of the Ethereum upgrades with forecast.org or we can uh track all the IPs the the timeline for the upgrades and everything is actually havinging to be on time as scaling L1. I think yesterday I was reading that we'll be maybe targeting 70 million barrier now and we'll see whether 100 million gas will be achievable maybe by March next year but there

context about 3x the capacity than than currently. So um that's that's a big improvement per for number of

comput computations you can run per block

and for everything else. So, so there are blog posts and and that's how we track it and then at the end March uh March June next year we'll start talking so much more about the next year of goals and we did a summary of the one year of efforts one year of focus we'll continue working on this but uh generally everyone is informed

everything's in the open I'll go a little bit deeper what's the preview of what the next year's goals might be I feel like you can never be done with scaling so how do you uh how do you think about like what's the next goal to uh we we mostly agreed that we'll be talking about finality about privacy and security. So we say we accelerate it uh but there are some very basic universal goals for Ethereum to always stay secure safe uh to continue delivering privacy and the finality is the one thing that we've heard the most uh to be talked about sometimes about finality.

Let me help people understand what finality is. So you have the

nontechnical include a few components. There will be shorter slot times most likely. there will be free slot finality. So the the time when you no longer for sure reverse transactions even in uh in catastrophic failures or very significant events of the on the consensus side. So we want to not only improve it but also clarify it.

So much of it on the finality will be education and uh clear definitions between L2s and L1's so everyone can very predictably judge which L2 provides what finality guarantees and uh this will be very very likely one of the major focuses uh for for all core devs for the researchers next year.

That's amazing. Can can I just uh point out that all of the goals of the Ethereum Foundation and the Ethereum community um are all under the umbrella of the overarching goal which is to maintain rigorous decentralization to do it right to do it right the first time. That's hard. That takes a long time or longer sometimes. But uh

I think the other nuance is that you're also never done like it's you can continuously make something better. So, we're we're building something extensible, infinitely extensible, horizontally, vertically, sideways, in high dimensions, who knows how how that's going to go. Um, and because you've got this ever expanding surface area. um you're going to either open up edge cases, vulnerabilities, you're going to open up opportunities for actors to come in and captize things and and capture and consolidate power. So, we always have to be vigilant against uh those sorts of threats.

That's amazing. um Shriram for kind of how Ethereum started as a world computer. We've kind of seen a big shift in how we think about comput al together especially with AI and I'm kind of curious to see how you think about the world moving forward and where does kind of blockchain come into this uh equation.

Yeah, you know um our thesis is always that uh Ethereum is the trust layer for humanity, right? So when you want to coordinate, you want to interact with each other, you need something to assure that you know that the outcomes of those interactions are going to be uh predictable, reliable, immutable, all these properties that you want from these systems. So when you think about where trust plays a role in the upcoming AI agentic era, you see basically trust breakdown in every single kind of like uh strata, right? So for for example yesterday Sora released the video models. Now I can take Joe Loin and like put him in in a video that know he he has no idea what is going on and it looks very real.

So this is just the starting point of like this massive acceleration we're going to see in the AI era. How do I trust my counterparty? How do I know that uh an AI agent does what it says it does? How do I know uh that a system how do I hold an AI to account? How do how do you hold these systems accountable?

And I think these are all the questions for which I think we should be building you know and extending Ethereum. So when you think about the journey of the world computer, the original Ethereum vision, what we see today has already happened or is tending to happen is the world ledger, right? It's a ledger where like we're all going to record, you know, uh and we see the convergence not just from our own crypto world but also from the rest of the banking world where people are interested in know settling and you know bringing Ethereum as the ledger of record. Um but the ledger is only one aspect of trust. You know the computations like an AI executed correctly or a game executed correctly and then brought the results back on chain.

These kinds of things are necessary and I think that's what the original world computer vision encompasses. What we've learned is to do this we need a scale of computation that goes far beyond what the Ethereum base layer can itself do. And so this was one of our original vision in building IGEN layer. You know Joe mentioned the infinite extensibility of Ethereum and layer is one of the best examples of this property. You can come in and stake ETH and then secure not just the Ethereum network but any other network that you can dream of.

Any decentralized verifiable network can be built on top of this. And so that that was the original idea of IGEN layer. IEN in German for your own your own layer. you can think of whatever you want to build and go and express that decentralized protocol on top of layer and in this AI world um you know we just launched actually two days back these pro products called IGEN compute and AI these are two products that basically allow uh securing of arbitrary offchain computation and arbitrary AI execution

while still preserving determinism that's correct so when you think about AI one of the biggest pro problems that you have today is you go and interact with it multiple times, you don't even get the same answer there. Some of it is by design. You want it actually to have some randomness. But there should be a way to control it where if you give the same random seed, you should get the same results. And so existing AI workflows and backends are not like that.

And so part of what we did with AAI is to produce a fully deterministic AI inference pipeline. So if you go query to it multiple times, you'll get the same answer. Once you make this deterministic, you can actually build accountability mechanisms for it. Because if a single node comes in and stakes ETH and promises to run your AI execution correctly and doesn't run it, you can actually find it out because it's deterministic. You can re-execute it and check that it wasn't done correctly and penalize them.

So this is part of our new uh new primitives I think which are needed to unlock and bring trust to the AI and agentic era. Amazing. Um it's similar to kind of what you what you just said, but I actually want to go down one one layer um sort of below this, which is understanding kind of how should people think about L2s now. Um you have a very successful layer 2 called Lineia and EF is spending a lot of time and effort on the engineering to scale at layer 1. Um how do we kind of see that intersection over the next couple of years?

Should people just think about building on L1's directly? Should we still think L2s have their own place? How do the economics change? Kind of how do you both think about these from from each of your perspectives?

So I can start and love to hear what what you say also. Um so we set out to um to build a new kind of meta uh in our L2. Um we had the benefit of looking at uh at the different L2s, optimistic L2s, some zero knowledge L2s. Um, we believe we we built uh the best zero knowledge EVM technology because we've got just maybe the best cryptographers, some of the best cryptographers in the space. Um, uh, but what we did was we recognized yes, we needed to fragment the ecosystem, need to fragment liquidity.

Um we the Ethereum layer 1 needed to seed pricing power to the layer twos to enable them to establish themselves and flourish. That's the the rollup ccentric scalability roadmap. Uh I think that has worked beautifully. Uh so that was a divergent explorative phase. Now we're in this conversion phase where where we figure out how to wire everything up so that's more homogeneous and mo more coherent and you don't really need to know where your token is, which network your token is on and and so we'll get that done with wallets and bridging systems on top

and yeah and intentric lots of things. Um most importantly I think for Lineia uh was that we uh decided that we needed to a be 100% compatible uh with Ethereum. Uh so all the all the consent all the execution clients uh sync to our network. Um same bite code basically which is not not true of the other rollups. Um and maybe most importantly um we wanted to support the mothership that uh that effectively enabled us to exist.

And so uh Ether is is the token on the network and we're burning uh 20% of the fees of every transaction and we're we're burning the rest of the fees uh in terms of linear burning. Uh and so um we are hoping we're not hoping that that remains a competitive advantage for us. We're hoping to set an example. Um we're also hoping that because we have 15% of the tokens allocated to consensus, the company that that built the thing, but we managed to convince our investors and our employees that we will all be better off if nobody else gets any tokens. Um except except about their hope, right?

So, so it gives us 85% of the tokens um to feed into the ecosystem to grant to builders, to invest in builders, to grant for, uh liquidity provision, to incentivize, um usage of the network. And so, um we're sitting at around $2 billion fully diluted value right now. So, we've got

84% 1.6 billion ecosystem development fund. And we are explicitly one. Yeah.

Yeah. We're explicitly going to fund projects that are linear aligned or Ethereum aligned. Um and uh um this is a way to to start we're actually starting out at a at a base where there's um it's a good valuation. There's no sell pressure going forward. So it's a rock solid base and and we think we're going to be able to grow a really sturdy token economy that way.

It's nice that you say that like uh L2 willing to support mainet the matter ship you say

crazy idea isn't it

at the same time the matter ship wants to support LDL2 so this is uh this is permissionlessness uh in working properly as planned so we want as many L2s as possible uh with the customization with the focus on particular region the focus on particular uh niche in the market u some of them targeting more global approach uh all the different kinds of innovation happening through the L2s and that's how people expressing either limitations specific to their business or or the new opportunities new technology experimentations uh but it doesn't mean that we don't want to continue uh improving the L1 scaling L1 and we scale L1 also for the L2s to be able to thrive and through the L2s being able to thrive the L1 is also thriving uh economically as well a lot of uh lots of liquidity is flowing to Ethereum mainet because uh that liquidity knows that it's it's sparked there to uh to continuously try the higher risk uh explorations of the L2s uh going to different markets and going to safe heaven to haven to Ethereum mainet and uh and we see it in numbers. So each time the L2s are succeeding, each time new L2s are launching, uh we see now the the per L2s, the high highly scalable L2s, uh all the different solutions, LINA exchanges launching uh in Korea, here in uh in Singapore, in Hong Kong. Uh all of this improves Ethereum ecosystem and we care a lot about the interoperability solutions that really benefits the entire ecosystem and it's it's a lot of talking of uh ensuring that everyone feels very comfortable that that future together uh is assured by not only by technology it's there right so it's permissionless so technology assures that but also by all the people building they they want to feel comfortable that we here together for many years forward uh believing together in that philosophy of the L2s at matters and yeah I'm particularly for this year uh I'm happy that we quickly discarded any questions about uh how the L21 collaboration works uh set it as part of the priorities for the protocol and going very strong and is seeing results of this now

amazing

uh we got

uh Karthik I'll just add one thing there on the L2s uh there is the scaling of computation and execution which is what the L2s are focusing on there is also the scaling of the data throughput you know that's also done on the L1 itself through blobs but we've built IGN DA as a mechanism to get out of protocol scaling of data throughput DA runs at 100 megabytes per second which is 1,00x what blobs run at today not just that it is 10x more throughput than all chains put together so you can actually build tens of solas on top of DA using you know as an Ethereum L2. So the scope of the the this you know dance between the L1 scaling and out of protocol solutions like layer DA rollups scaling helping support each other I think is a really amazing thing happening in the Ethereum on this one. What do you think uh Vitalik's role should be for the next 10 years to help scale Ethereum?

I think Vitalik's role is the kind of

four more words left.

Sorry,

four more words left. You got to be careful.

Four more words.

Uh be the visionary and the values holder of Ethereum.

He wants to work on hardware like he he mentioned that uh last days. So hardware engineering, verification of hardware,

whatever the hell he wants to do, hopefully diverse.

Amazing.

And my last question and uh you can have a single sentence on this one. Um obviously people are are choosing between different ecosystems, different chains. We know what the road map looks like for uh the scalability of Ethereum. So, as a founder who's thinking about building something um on Ethereum, what advice would you give them um as they consider which chain to pick to spend the next 10 years of their lives? I'll start with you.

Um so, I would say don't worry about which chain. We're going to solve that. Uh everything will be abstracted and and you should just focus on what you care about. Um get expert in it or if you're a generalist, uh you know, pick pick the things pick the topics that you care about. There's no better time to be coming of age or coming into the market or doing a career change than at the end of a super cycle when everything's falling apart because uh because everything is going to be fresh.

There's going to be a new social contract, a new system of the world and uh and you're going to be able to build it.

Um it's very easy to move between the chains on Ethereum. So wherever they feel the most supported, the more most comfortable, most excited about the branding, they should go there. The infra is ready, be much, much more ambitious. The four apps we have in crypto are not enough. We came here to disrupt the entire economy and we will not stop till we get there.

Parting advice, pick the ones with the best values that resonate with you. And with that, Shriram, Tamas, Joe, thank you so much for an amazing conversation.

Thank you so much.

There we go.