Exclusive Interview with Ethereum Founder Vitalik: Interpreting POS Upgrade, L2 Development, and the Integration of Cryptography
On April 7, 2025, at the Pop-X HK Research House event held in Hong Kong, Ethereum founder Vitalik Buterin accepted a Chinese interview, sharing his insights on several key topics related to the Ethereum ecosystem. Vitalik answered various hot topics, including ETH POS, Layer2, Cryptography, and AI, in fluent Chinese.
When asked about his views on Ethereum's POS upgrade, Vitalik stated: "I believe that the prices of BTC and ETH have nothing to do with POW and POS. The things that the BTC and ETH communities are doing are completely different, and their ways of thinking are also completely different."
He pointed out a key problem: "There will be a lot of successful applications on Ethereum, but these successful applications may not bring enough value to ETH." "It's similar to Google, where 90% of its revenue comes from the search business, despite its many products.
"We need to support more applications that have long-term value for Ethereum holders and ETH," Vitalik emphasized. ETH's future success is likely to occur in these areas, and it doesn't have much correlation with improvements in consensus algorithms. "
02 PBS Architecture and Centralization Concerns
Vitalik analyzed the potential issue of centralization in the Builder under the PBS architecture for ETH2.0 from a philosophical perspective.
He pointed out that an early subconscious assumption of Bitcoin and Ethereum was "building blocks and validating blocks are one operation," but now with new technologies like ZK and DAS, the two can become asymmetric: "the difficulty of building a block may become very high, but the difficulty of validating a block may become very low."
Regarding the risk of supernode evil, Vitalik analyzed three possible attack methods: 51% attack, censorship attack, and MEV-related operations. He explained that since the verification process is done through Attester, the cost of verification is low, so the barrier to entry for consensus nodes is still low. Even if the supernode controls most of the build power, it cannot interfere with the entire consensus process.
He also mentioned the concept of Fossil: "Fossil is the ability to separate the role of 'selecting transactions' from the role of executing them. This way, the role of choosing which transactions are included in the next block can be made more decentralized."
03 The Relationship and Future Direction of Layer 1 and Layer 2
When discussing the relationship between Layer 1 and Layer 2, Vitalik believes that there are several issues with the current Layer 2.
First of all, Layer 2 is not progressing fast enough in terms of security, and he has been pushing for Layer 2 to be upgraded to Stage 1, hoping to upgrade to Stage 2 this year.
The second is the issue of interoperability between Layer 2s: "Perhaps 80% of L2 cross-chain problems are actually UX problems. To this end, they started the Open Intents Framework and Chain-specific addresses projects.
He also supports a 2 of 3 verification method (OP + ZK + TEE), which fulfills three requirements: complete trustlessness, starting with ZK technology, reducing withdrawal time from one week to 1 hour. The long-term plan is to further reduce the withdrawal time to 12 seconds, and even to 4 seconds with SSF.
Regarding Layer 1's role, Vitalik said, "If L1 doesn't play any role at all, it's dangerous for ETH. One of our biggest concerns is that the success of Ethereum applications cannot be the success of ETH. "
He proposed that some high-value applications, such as DeFi, are more suitable for running on L1, especially those with longer investment periods. Therefore, they are working to enhance the scalability of L1 and plan to increase the Gas limit of L1 by ten times by 2026 through four to five methods (including delayed execution, block-level access lists, multidimensional gas pricing, and historical data processing).
Regarding the recent upgrades, Vitalik revealed: "The Pectra upgrade will take place at the end of this month, followed by the Fusaka upgrade in Q3 or Q4."
04 Cryptography and the Development of AI
In terms of cryptography, Vitalik believes that developers should be tiered and not need to understand all the underlying details: "Most ZK developers can know some high-level languages (HLL), write application code in HLL, while those researchers of the Proof System can continue to improve the underlying algorithms."
He pointed out that although the Circom and Groth16 ecosystems are developed, they have limitations, such as the need to set up their own trustworthiness and inefficiency. The ZK RISC-V route is a good direction because RISC-V can become an HLL on which many applications, including EVMs, can be developed.
Regarding the application of FHE and MPC, Vitalik expressed caution: "MPC and FHE always need a committee, and if the node is attacked, the system will have problems. And these nodes need to be online all the time, making the attack simpler. "
Speaking about AI, Vitalik shared a personal experience: "AI can sometimes be said to have a democratizing effect. When I use AI myself, in those fields where I am already in the top 1,000 in the world, such as ZK development, AI helps less; But in the field where I'm a rookie, AI is very helpful. The gap between experts and rookies has been lowered a lot with the help of AI. "
05 Expectations for the Ethereum Ecosystem
Finally, Vitalik expressed expectations and suggestions to the developer community:
"There are many opportunities for developing applications on Ethereum now, and many things that were previously impossible can now be done. The TPS issue of L1 has been resolved, privacy issues have solutions, and AI has reduced the difficulty of development."
He pointed out the biggest problem facing the current blockchain application ecosystem: "We have two types of applications, one is very open, decentralized, secure, and particularly idealistic, but it only has 42 users; the other is a casino. Both extremes are unhealthy."
Vitalik hopes to see applications that: first, have real value and users enjoy using; second, are better for the world; and third, have a sustainable business model.
"Now everyone has more resources than before, and if you can find a good idea and execute it well, the chances of success are very high."
The content is for reference only, not a solicitation or offer. No investment, tax, or legal advice provided. See Disclaimer for more risks disclosure.
Vitalik answered in Chinese throughout: Ethereum POS upgrade, Layer 2 challenges, and the way of capturing ETH value.
Exclusive Interview with Ethereum Founder Vitalik: Interpreting POS Upgrade, L2 Development, and the Integration of Cryptography
On April 7, 2025, at the Pop-X HK Research House event held in Hong Kong, Ethereum founder Vitalik Buterin accepted a Chinese interview, sharing his insights on several key topics related to the Ethereum ecosystem. Vitalik answered various hot topics, including ETH POS, Layer2, Cryptography, and AI, in fluent Chinese.
! Chinese Community Conversation Vitalik: POS, L2, Ethereum and AI Convergence Innovation
01 Views on POS Upgrade
When asked about his views on Ethereum's POS upgrade, Vitalik stated: "I believe that the prices of BTC and ETH have nothing to do with POW and POS. The things that the BTC and ETH communities are doing are completely different, and their ways of thinking are also completely different."
He pointed out a key problem: "There will be a lot of successful applications on Ethereum, but these successful applications may not bring enough value to ETH." "It's similar to Google, where 90% of its revenue comes from the search business, despite its many products.
"We need to support more applications that have long-term value for Ethereum holders and ETH," Vitalik emphasized. ETH's future success is likely to occur in these areas, and it doesn't have much correlation with improvements in consensus algorithms. "
02 PBS Architecture and Centralization Concerns
Vitalik analyzed the potential issue of centralization in the Builder under the PBS architecture for ETH2.0 from a philosophical perspective.
He pointed out that an early subconscious assumption of Bitcoin and Ethereum was "building blocks and validating blocks are one operation," but now with new technologies like ZK and DAS, the two can become asymmetric: "the difficulty of building a block may become very high, but the difficulty of validating a block may become very low."
Regarding the risk of supernode evil, Vitalik analyzed three possible attack methods: 51% attack, censorship attack, and MEV-related operations. He explained that since the verification process is done through Attester, the cost of verification is low, so the barrier to entry for consensus nodes is still low. Even if the supernode controls most of the build power, it cannot interfere with the entire consensus process.
He also mentioned the concept of Fossil: "Fossil is the ability to separate the role of 'selecting transactions' from the role of executing them. This way, the role of choosing which transactions are included in the next block can be made more decentralized."
03 The Relationship and Future Direction of Layer 1 and Layer 2
When discussing the relationship between Layer 1 and Layer 2, Vitalik believes that there are several issues with the current Layer 2.
First of all, Layer 2 is not progressing fast enough in terms of security, and he has been pushing for Layer 2 to be upgraded to Stage 1, hoping to upgrade to Stage 2 this year.
The second is the issue of interoperability between Layer 2s: "Perhaps 80% of L2 cross-chain problems are actually UX problems. To this end, they started the Open Intents Framework and Chain-specific addresses projects.
He also supports a 2 of 3 verification method (OP + ZK + TEE), which fulfills three requirements: complete trustlessness, starting with ZK technology, reducing withdrawal time from one week to 1 hour. The long-term plan is to further reduce the withdrawal time to 12 seconds, and even to 4 seconds with SSF.
Regarding Layer 1's role, Vitalik said, "If L1 doesn't play any role at all, it's dangerous for ETH. One of our biggest concerns is that the success of Ethereum applications cannot be the success of ETH. "
He proposed that some high-value applications, such as DeFi, are more suitable for running on L1, especially those with longer investment periods. Therefore, they are working to enhance the scalability of L1 and plan to increase the Gas limit of L1 by ten times by 2026 through four to five methods (including delayed execution, block-level access lists, multidimensional gas pricing, and historical data processing).
Regarding the recent upgrades, Vitalik revealed: "The Pectra upgrade will take place at the end of this month, followed by the Fusaka upgrade in Q3 or Q4."
04 Cryptography and the Development of AI
In terms of cryptography, Vitalik believes that developers should be tiered and not need to understand all the underlying details: "Most ZK developers can know some high-level languages (HLL), write application code in HLL, while those researchers of the Proof System can continue to improve the underlying algorithms."
He pointed out that although the Circom and Groth16 ecosystems are developed, they have limitations, such as the need to set up their own trustworthiness and inefficiency. The ZK RISC-V route is a good direction because RISC-V can become an HLL on which many applications, including EVMs, can be developed.
Regarding the application of FHE and MPC, Vitalik expressed caution: "MPC and FHE always need a committee, and if the node is attacked, the system will have problems. And these nodes need to be online all the time, making the attack simpler. "
Speaking about AI, Vitalik shared a personal experience: "AI can sometimes be said to have a democratizing effect. When I use AI myself, in those fields where I am already in the top 1,000 in the world, such as ZK development, AI helps less; But in the field where I'm a rookie, AI is very helpful. The gap between experts and rookies has been lowered a lot with the help of AI. "
05 Expectations for the Ethereum Ecosystem
Finally, Vitalik expressed expectations and suggestions to the developer community:
"There are many opportunities for developing applications on Ethereum now, and many things that were previously impossible can now be done. The TPS issue of L1 has been resolved, privacy issues have solutions, and AI has reduced the difficulty of development."
He pointed out the biggest problem facing the current blockchain application ecosystem: "We have two types of applications, one is very open, decentralized, secure, and particularly idealistic, but it only has 42 users; the other is a casino. Both extremes are unhealthy."
Vitalik hopes to see applications that: first, have real value and users enjoy using; second, are better for the world; and third, have a sustainable business model.
"Now everyone has more resources than before, and if you can find a good idea and execute it well, the chances of success are very high."