Conversation with Vitalik Buterin: The Cancun upgrade is technically progressing smoothly, re-staking is interesting but there are systemic risks

Author: Weilin, PANews

On March 21, Ethereum founder Vitalik Buterin participated in ETHTaipei 2024 and was interviewed by PANews and other media, talking about Cancun upgrades, re-staking and AI Crypto and many other hotly discussed topics.

He believes that from a technical point of view, the Cancun upgrade is going smoothly, but after the upgrade, as the number of blob users increases, network fees are likely to increase again. He is optimistic about the broad impact that stablecoins and decentralized social networking on Ethereum will have on society in the future. In addition, when talking about the recent MEME coin, Vitalik also thought it surprised him.

In the field of AI Crypto, one of the biggest challenges currently is bugs in our code. One possibility is that AI allows us to more easily use verification tools to prove the correctness of a larger range of code. By reducing the number of errors we can reduce, the space becomes safer. In addition, the re-staking track is interesting, but if not done well, various systemic risks may eventually affect Ethereum’s validators.

The following is an edited transcript of the conversation.

1. How does the Dencun upgrade contribute to the Ethereum ecosystem? Can you share how EIP-4844’s reduction in transaction costs will affect the Ethereum ecosystem?

Dencent’s upgrades are designed to significantly increase its scalability while reducing transaction fees paid by Layer 2 networks (especially rollups). At the core of the upgrade is the introduction of an independent portion of the data space called a "blob", one contained within each block. This data is invisible to the EVM, which is crucial because it means clients do not need to access this data at the same time when validating Ethereum blocks.

But these data are data guaranteed to be available on the Ethereum blockchain. This is useful for rollups, but also for any Layer 2 project that relies on data availability for security. For example, making sure other nodes are able to sync, if the current node disappears, or if someone sees the wrong answer, make sure they can challenge it, etc. We've seen L2 network fees drop significantly over the past week, in some cases as much as 50%.

Now I think it is necessary to remind you that as the number of blob users increases, these fees are likely to increase again. From a technical perspective, the upgrade went well. I think the percentage of testers only dropped from 99% to 95%, which is better than any of our previous forks. Interestingly, however, usage is relatively low. The target is about three blobs per block, but average usage is only one blob per block, which means blobs are currently very cheap. I think usage will eventually increase over time. Rollups are very cheap at the moment, which is definitely a good thing.

#2. What do you think Ethereum will have the most transformative impact on society in the next five years?

The next five years will be crucial for Ethereum, as many applications, although at a fairly theoretical and small-scale stage, are actually starting to become ready for real-world use.

One of the major real-world impacts in the blockchain field is that the ideas generated in this field have penetrated into the wider world in many ways and been recognized. For example, Reddit is about to IPO. One of the things they do is give very active contributors and moderators, etc., who are active in the community, the same opportunity to participate as institutional investors.

But beyond that, the biggest impact on actual use is stablecoins. People use stablecoins to store and trade. What do I think will happen in the next five years? First of all, to be successful in this space, you need to have a very good user experience and very low fees. Historically, Ethereum has not had these things, but in the next five years, if Ethereum starts to have these things, we have seen the L2 network already starting to move in this direction. I think this improvement will continue.

Once that is achieved, I expect Ethereum to become the dominant player and help make stablecoins acceptable in a way that is actually open, that is actually decentralized. Beyond that, I expect non-financial applications will really start to have a bigger impact. So, over the past year, we've looked at the success of Farcaster and particularly to some extent Lens and a few other projects that are creating alternatives to social media platforms.

The Ethereum-based identity space is something I expect to grow rapidly and the technology is improving rapidly.

3. We have seen many projects proposing the idea of ​​"modular blockchain", and we have also seen the idea that Ethereum L1 may be responsible for shared sequencing. What do you think, which functions should be handled centrally by L1, and which functions should be left to individual L2s?

The emergence of modular blockchains basically points to a future where each individual chain completes fewer tasks, and different components are completed by different parts.

If you look at the tasks that Layer1 is currently responsible for, L1 is currently responsible for shared security. It is a shared settlement layer. Basically, with Ethereum, it ensures that every L2 is able to read any other L2 in a way that does not rely on any centralized actor, or even any valid data set. Because even if Ethereum suffers a 51% attack, all L2 will be restored at the same time and you will still maintain consistency.

The question is, what is the ideal way forward? I personally am neutral on the issue of shared ranking. For other types of functionality, I think we definitely want to expand the amount of data supported directly by L2 or Ethereum. In an ideal world, everything would be a rollup and Ethereum wouldn't handle all data availability. But we know that in practice, this may not be enough, and even the long-term vision of 16 Mbytes/slot may not be enough. What I want to see is I want very high security stuff to come out of on-chain rollups.

4. In a tweet in February, you discussed the role of AI in debugging code, which is crucial to Ethereum. Can you think further about how AI is used in cryptocurrencies and Ethereum?

Over the past 10 years, many people have been very curious about the intersection of artificial intelligence and cryptocurrency. But the question has always been, can this thematic fusion result in practical application examples that can rationally utilize both in production? In my latest article, published about two months ago, I tried to analyze this problem and tried to identify some specific application scenarios.

One of them is the participation of AI in prediction markets or other types of markets on Ethereum, which can make the market more micro and then allow AI to participate in it. Another is AI as part of the wallet, helping users understand the online and on-chain environments they interact with. The third is to use cryptography to try to create AI models that are secure, powerful and privacy-preserving, so that they can be safely used as a central participant in on-chain applications, whether it is a DAO or other applications such as oracles.

Then the fourth is that if it is successful, then it can be used for AI in other fields. Of these applications, I think the first two are the most obvious and short-term, while the latter two are actually more speculative. So I definitely don't want to give the impression that AI crypto applications are going to be the next big narrative driving the industry or whatever. But I definitely think these intersections are worthy of people looking into.

The other is the role of AI in code debugging. Basically, one of the biggest challenges facing our field right now is bugs in our code. Perhaps one of the good possibilities we can look forward to is that AI will make it easier for us to use verification tools to prove that a larger range of code satisfies certain properties. If we do this, then it is possible for us to have guaranteed bug-free code like the ZK EVM. By reducing the number of errors we can reduce, the space becomes safer. AI could be very valuable in this regard.

#5. What do you think of the re-staking craze in Ethereum?

I think restaking is interesting because on the one hand it has the potential to be a very valuable way to unlock funds that have been staked and make them available to other applications Available.

Like this is one of those things where if there's a need and we're not filling that need in some way, there's basically a risk of that need being taken over by centralized actions . Because anyone can accept other people’s Ethereum and stake it on their behalf, then issue tokens and then have people re-stake those tokens. But this is not the world we want to see. But on the other hand, there are indeed risks in re-staking. If it is not done well, various systemic risks, such as various systemic risks, will eventually affect the validators of Ethereum.

So there are a few different approaches to this problem. I know there are a few different projects that have been being worked on. So far, I'm just waiting to see what comes out of this space.

6. Are you more active on Farcaster than on Twitter?

An interesting thing about the social field is that it has network effects, but it also has anti-network effects. It's like Twitter is where everyone is, but it's also where all the really annoying people are.

So far, I've found Farcaster to a point where it's interesting enough that there are enough people there, but at the same time, the quality of interactions I get there is actually higher.

Another important benefit that decentralized social platforms can really bring is that one of the main issues in social platform management, transformation and all discussions around this topic in the past few years is that Twitter has and uses some Function, basically, to differentiate between high-quality content and people, and low-quality content and people. And in the cryptocurrency space, I think the fundamental purpose of a lot of these non-financial Ethereum applications is to solve this problem in a more decentralized way and try to solve the trust problem.

I think these things may eventually be combined with Farcaster, as well as Lens and some other protocols. We may actually see a very vivid example of trying to see if these techniques work. Farcaster has definitely started to have real spam problems. So that's something they need to solve. But I'm glad to see this.

7. What has been most surprising to you about the development and evolution of cryptocurrencies? For example, what would you tell yourself 5 years ago?

First of all, in 2020, as I may have said elsewhere, NFTs really surprised me. I think it's really surprising that a popular meme coin is popping up again right now. On the positive side, I think the rapid progress of ZK technology does surprise me. This is something that happened faster than we expected, which is very rare in the software world.

I think the second part is that level one protocol changes are being rolled out faster and faster. Like we did the merger and then we did the blob and this was only 1.5 years after the merger, right? And it looks like we will complete Verkle trees in the near future. So I was impressed with that as well.

The above is the detailed content of Conversation with Vitalik Buterin: The Cancun upgrade is technically progressing smoothly, re-staking is interesting but there are systemic risks. For more information, please follow other related articles on the PHP Chinese website!