Exclusive Interview with Vitalik: From World Computer to World Ledger, Where Is Ethereum Heading in the Next Decade?
TechFlow Selected TechFlow Selected
Exclusive Interview with Vitalik: From World Computer to World Ledger, Where Is Ethereum Heading in the Next Decade?
"The greatest value of Ethereum lies in its openness."
Navigating Web3 tides with focused insightsCompilation & Translation: TechFlow
Guest: Vitalik Buterin, Co-founder of Ethereum
Hosts: Ryan Sean Adams; David Hoffman
Podcast Source: Bankless
Original Title: Vitalik Buterin: How Ethereum Becomes The World Ledger
Release Date: August 11, 2025
Key Takeaways
Ethereum has reached its tenth birthday. Founder Vitalik Buterin reflects on the network's decade-long journey and shares his vision for the future. He discusses unexpected surprises and challenges along the way—such as The DAO and NFTs—and what he might have done differently if starting over.
This conversation dives deep into the evolution of Ethereum’s culture, the importance of privacy as a core value, and the trade-offs between L1 and L2. Vitalik also shares his thoughts on how Ethereum can evolve in an AI-driven future.
Highlights Summary
-
One of Ethereum’s greatest contributions has been advancing openness and decentralization, making these values second nature to many. These ideals must be passed down and renewed with each generation.
-
If I could go back in time, I would give my younger self all the knowledge we now have about zero-knowledge proofs (ZK-SNARKs).
-
Ethereum should be an inclusive ecosystem that welcomes diverse people and perspectives.
-
The concept of Ethereum as a “world ledger” is more concrete and better communicates its core value.
-
The emergence of NFTs was a huge surprise. I never expected them to happen at all.
-
If we think of the ledger (Ethereum) as a book, then ETH is like the "ink" used to write it. That ink can represent L2 functionality. If the ink has a TM (trademark symbol), it means it's a branded L2; without TM, it's an unbranded L2.
-
Every project has a responsibility to build the best version of the philosophy its members are most passionate about. The goal should be creating things beneficial to the world—not endless infighting or shouting slogans.
-
Privacy is essential—it is freedom, a right we must all protect. Everyone in the Ethereum ecosystem should support privacy.
-
We shouldn’t have “privacy wallets.” Privacy should be a feature of all wallets, seamlessly integrated into existing ones.
-
A bloated centralized data collection system is fragile. If such databases are hacked, data thought to aid national security could actually destabilize it.
-
Ethereum’s role has two parts: first, providing tools that protect people’s freedom, autonomy, and organizational capacity—tools independent of any individual, company, or nation; second, building a global community. Ethereum attracts those who care about decentralized finance, novel organizational forms, privacy, and democratic governance—an irreplaceable community asset.
-
The difficulties of 2024 stem partly from ETH’s price decline pressuring the ecosystem, and partly from once-popular projects and topics fading while new ones haven’t yet emerged. The path forward must involve new projects that deliver tangible benefits and broad support.
-
L1 should have moderately low latency to meet average user needs; for ultra-low-latency use cases like high-frequency trading, L2 is better suited. We need L1 and L2 to work together for balanced ecosystem development.
-
Ethereum’s greatest value lies in its openness. It’s like a sandbox enabling experimentation and innovation in many directions.
-
“I believe participants in the Ethereum ecosystem, including those involved in financial activities, are responsible and won’t take reckless risks. ETH derivatives are a fundamental form of financial stability, so I support the existence of these treasury companies.”
-
Blockchain technology can help create a fairer, more transparent world.
The First Two Years of Ethereum
David:
Ethereum just celebrated its tenth anniversary. First, happy birthday to Ethereum! The whitepaper was first published in 2013, and the mainnet officially launched in July 2015.
Ten years later, Vitalik, how does Ethereum’s development compare to your original vision?
Vitalik:
In the most important way, Ethereum has far exceeded my expectations. At the same time, the process has taken much longer than I anticipated.
When I wrote the whitepaper, I planned it as a side project to finish in a few months before returning to university. That didn’t happen. Later, we envisioned Ethereum going through four stages before transitioning fully to Proof of Stake. We thought the foundation would run out of funds and the project would naturally become self-sustaining—but that didn’t happen either.
Also, the rise of decentralized finance (DeFi) and various tokens were applications mentioned in the whitepaper. Though the terminology people eventually used differed from mine, these developments did materialize—like Ethereum Name Service (ENS) and stablecoins. And there were many surprises too.
Ethereum’s Major Contributions
David:
Ethereum has now completed ten years. Looking back, many things unfolded largely as planned. If you read the original whitepaper and compare it to today’s Ethereum, you’ll see it fulfilled many early visions. Reflecting on this history, what unique contributions do you think Ethereum has made to the world? What makes you most proud or satisfied?
Vitalik:
I think one of Ethereum’s greatest contributions has been advancing openness and decentralization, making these values second nature to many. These ideals must be passed down and renewed with each generation. Just as free and open-source software rose in the 1980s and 1990s, the blockchain world inherited and extended that spirit into the 2010s and 2020s. Ethereum has notably turned theoretical ideas into reality.
For example, prediction markets were purely theoretical in the early 2010s—a mechanism for analyzing event outcomes. Ethereum provided a key experimental platform, turning this idea into practice.
Another example is DAOs (decentralized autonomous organizations). Despite many setbacks, Ethereum made blockchain-based governance more flexible and actionable. I’m very proud of this, and I believe we’ll continue seeing progress in these areas for decades. Ethereum’s impact isn’t limited to one domain—it’s had profound influence across technology and social philosophy.
Biggest Surprises
Ryan:
You mentioned some surprises along the way. There must have been many unexpected events. Interestingly, you initially thought Ethereum would be a side project, but it became a full-time endeavor lasting over a decade.
What were the biggest surprises during Ethereum’s development?
Vitalik:
First, the sheer amount of ETH that poured into the DAO, followed by its near-immediate collapse. In hindsight, the massive funding was even more surprising than the crash, though both were striking.
Ryan:
It was like an uncontrollable use case—or an early instance of DeFi, since it involved capital formation. I remember around 5% of the ETH supply went into the DAO, which is huge.
Vitalik:
It was actually 11%. Total ETH supply at the time was 11 million. A lighter version might have hit 17%. Either way, it was extremely high—the first time in Ethereum’s history.
Then there’s the birth of Ethereum Classic (ETC). The hard fork controversy felt like a war. It was almost like a TV drama. Right after ETC settled, the Shanghai DDoS attacks began—as if scripted. These brought technical challenges that were fascinating.
The emergence of NFTs was another huge surprise. I never expected NFTs to appear. The growth of DeFi surprised me too. In 2019, DeFi was tiny—barely Uniswap holding on. Within a year and a half, it exploded. Also, the transition to Proof of Stake (PoS) took much longer than expected—that’s worth reflecting on.
On the positive side, zero-knowledge proof (zk) technology advanced five times faster than I expected, which was very encouraging. Also, institutional and governmental interest in blockchain exceeded my expectations. Even in the 2010s, major corporations and governments showed strong interest—something I didn’t foresee. Back then, it was often for showcasing innovation, but still unexpected. Now, institutional interest is returning in more concrete, practical forms. Many other things surprised me too.
Things That Took Longer Than Expected
Ryan:
Vitalik, you mentioned a core issue: many things took longer than you initially thought, especially when you released the Ethereum whitepaper. Why did they take longer? When you say “took longer,” do you mean achieving Proof of Stake (PoS) and implementing the rollup roadmap? Which specific parts took longer than expected, and why?
Vitalik:
I think one major reason is the complexity of software development. At the time, I lacked experience and underestimated the difficulty. Another reason is that we kept raising our standards during development. Our initial plan was to launch a Layer 2 solution based on Prime Coin within months. But when we saw growing attention and interest in January, we realized Ethereum carried many people’s hopes—it deserved serious effort. So we decided to build a proper L1 capable of supporting Layer 2 solutions.
Back then, few L1s were suitable for building Layer 2s. Together, these factors prolonged technical development and pushed us to continually improve our standards.
Overcoming Challenges
David:
Drawing from Ethereum’s journey, you’ve faced many unpredictable challenges—like the DAO hard fork and the Shanghai attack. Yet even during Ethereum’s successes, such as the 2021 NFT boom, new challenges arose. The project has weathered complex crises.
How specifically did Ethereum develop strategies to handle these challenges? How did this strategy gradually form among the culture, community, foundation, and core developers? After nearly ten years of experience, how would you describe Ethereum’s unique approach to solving unforeseen challenges?
Vitalik:
I think our ecosystem-level problem-solving approach has been very effective. We always try multiple different approaches. For example, there are solutions built on the blockchain base layer (L1), and others developed directly at the application layer. Within each category, multiple competing solutions often exist simultaneously. This allows parallel progress and creates synergy between different efforts—especially in maturing zero-knowledge proof technology.
Additionally, I find Ethereum’s collaborative style impressive. It’s not perfect, but overall, it works very well.
Lessons for Younger Vitalik
David:
If you could go back in time and share knowledge with younger Vitalik or the newly formed Ethereum Foundation to help Ethereum develop better, which moment would you choose? What would you teach them?
Vitalik:
An obvious answer is going back to Ethereum’s earliest days and giving them all the knowledge we now have about zero-knowledge succinct non-interactive arguments of knowledge (ZK-SNARKs). This technology is disruptive and powerful in many ways.
David:
To give Ethereum’s zero-knowledge tech a ten-year head start?
Vitalik:
Exactly. I think we took many detours or unnecessary paths in tech development. Knowing the end goal earlier would have saved resources and accelerated progress. So not knowing the future is indeed a regrettable limitation in Ethereum’s evolution.
People often ask me what message I’d send to my past self. Often, my answer is simply pointing toward the technically correct direction. But I also wonder—beyond tech—are there other messages worth sending? Like advising myself to have more realistic timelines.
Sometimes I think, were there better social or economic strategies we could’ve adopted that would feel more ideal in hindsight? For example, in Ethereum’s early days (2014), could we have introduced a temporary mechanism to allocate part of the token supply to public goods projects? Even if implemented crudely—say, miner voting (based on the last 1024 blocks) to decide developer rewards. Such a mechanism might avoid explicit pre-mining, provide better funding for the foundation and other groups, and perhaps earn Ethereum more credibility early on.
I often wonder if Ethereum could’ve built stronger ties with the Bitcoin community early on. It’s a pity we didn’t attract more Bitcoin supporters.
Here’s a bold hypothesis: if Ethereum had used the formula I mentioned for token issuance, started as a Bitcoin fork, and announced a move to Proof of Stake from day one—even if the initial mechanism wasn’t perfect. Perhaps Ethereum could’ve joined Bitcoin’s “big block” faction—the group supporting larger block sizes to increase transaction capacity. Had Ethereum aligned with them, the entire journey might’ve been smoother.
Of course, any choice brings unintended consequences. Starting with an existing community might impose constraints from stakeholders, limiting flexibility. These questions deserve deep reflection.
Bitcoin vs Ethereum
Ryan:
I feel Bitcoin’s ideology or philosophy has evolved into something almost religious. This makes me wonder if any forked coin inevitably causes community fragmentation or even collapse. But over a decade has passed—Bitcoin is about 16 years old—so we’ve all grown up together. You could say Ethereum is in rapid adolescence, while Bitcoin is like a young adult just gaining independence.
Do you think the relationship between Bitcoin and Ethereum has improved? Or is it just changing due to price ratio shifts? I notice that when Bitcoin underperforms, its community seems friendlier toward Ethereum. Now they’re quieter, but attitudes may shift again with price trends. Still, I sense less hostility between newer Bitcoin and Ethereum communities. How would you describe this relationship?
Vitalik:
That depends on what you mean by “new generation,” because I see different types. Some focus on tech—researching VMs, Taproot (a privacy and efficiency upgrade for Bitcoin), OP_CAT (a script opcode to extend Bitcoin’s capabilities). Others, like followers of Michael Saylor, probably won’t ever be particularly friendly or aligned with Ethereum’s values.
Technically, smart people in the space do highly appreciate Ethereum’s technical advances and privacy efforts. These aren’t just theoretical—they’ve produced real results. Also, some in the Bitcoin community are trying to advance tech via OP_CAT and new Layer 2 solutions like Lightning Network—efforts I find very interesting. So technically, the Bitcoin-Ethereum relationship has become more positive.
David:
When you see people building on Bitcoin—trying to enhance itsVM capabilities or make it more expressive—do you think, “You’re wasting time—just come build on Ethereum! That’s exactly what we created it for”? Or do you view their attempts with curiosity and optimism?
Vitalik:
I feel a bit of both.
The Evolution of Ethereum’s Culture
David:
I want to discuss Ethereum’s current stage. Recently, following Donald Trump’s election, we’ve observed a cultural shift—not only in crypto but in broader society. In a sense, it reflects a move away from a feminized World Economic Forum (WEF) style toward a more traditional “Bronze Age mindset.” As one Twitter user put it, it’s an “out with woke, in with base” cultural trend. This zeitgeist has seeped into crypto, with many projects emphasizing American roots. For example, Solana’s controversial marketing video explicitly stated its focus on tech innovation, not gender issues.
(TechFlow note: “WEF style” refers to values promoted by the World Economic Forum—globalization, inclusivity, diversity—sometimes perceived as “feminine.” “Bronze Age mindset” is a metaphor for more traditional, primal values emphasizing strength, foundational building, and cultural tradition.)
Notably, Ethereum doesn’t seem to be joining this cultural shift. Your thoughts? Whether Ethereum was like this before or not, it remains unchanged today. Is this intentional? Should Ethereum serve as a bulwark against shifting tides?
Vitalik:
I think Ethereum should be an inclusive, pluralistic ecosystem that accommodates many different people and viewpoints. Yet in a world with many cryptocurrencies and ecosystems, natural cultural selection will occur. Even if Ethereum were the only cryptocurrency, cultural differences with other projects would persist.
I believe every project has a responsibility to build the best version of the philosophy its members are most passionate about. The goal should be creating things beneficial to the world—not endless infighting or shouting slogans. You see one group shouting one slogan, another shouting a different one, feeling righteous—yet months later, nothing has changed.
Personally, I am concerned about the intense cultural shift we’re seeing. But if you only worry, you just become part of it. So the real question is: how do we respond, advance, and create a better alternative?
One new theme this year has been my attempt to revive discussions around DAO public goods funding. I think these topics are vital. Abandoning public goods funding or any governance beyond founder dictatorship could lead to negative outcomes—eventually reverting to reliance on founder authority.
But quadratic funding has issues, and tokenvoting DAOs face many challenges. So I’ve supported prediction market-based funding and prediction market DAOs. I collaborate with Divonch, who’s done extensive work in public goods funding. Version two is prediction-market-based—essentially combining prediction markets with a jury mechanism. The idea is scaling value assessment via prediction markets, while using juries to ensure high-quality evaluation.
My basic philosophy is replicating the best aspects offree markets in public goods funding, creating an open participation order. Such a system lets anyone join—if they perform well, they get a fair chance, rather than devolving into a pure social game.
Another priority I’ve pushed is greater focus on privacy. Privacy has always been central to cypherpunk ideals. Recall Chaum’s e-cash in 1982—it wasn’t decentralized, relying on a central operator, but it was private. Due to tech limits, we moved to Bitcoin—decentralized but not private. Now with ZK-SNARKs, those limits are gone—we can be both decentralized and private.
Thus, I believe privacy is essential—privacy is freedom, a right we must all protect. We must build privacy into the technology. If privacy is our focus, we’re playing a game of action, not talk.
I believe everyone in the Ethereum ecosystem should support privacy. We must keep pushing in this direction.
Privacy on Ethereum
Ryan:
Let’s dive deeper, because I think privacy is an interesting and important part of Ethereum’s culture—balancing core values with mainstream adoption. You mentioned Milady, which seems to embody “Bronze Age pop”—front-end is Bronze Age imagery, back-end is effective policy and real work. So how do we achieve “Bronze Age pop” for privacy on Ethereum?
In some areas, we’ve made huge cryptographic advances—like amazing ZK tech. But in practical privacy applications like Tornado Cash, we still face challenges. Roman Storm’s trial is ongoing—we don’t know the final verdict. States often intervene when financial privacy is involved.
I’ve long believed that if Ethereum or Bitcoin had privacy from the start—all transactions private—we might not have gained widespread acceptance. Forces might have tried to crush it before the tech matured.
So how do we strike a balance—ensuring privacy develops reasonably without provoking strong regulatory backlash? Currently, privacy apps like Tornado Cash and Railgun remain niche, UX is poor, and privacy isn’t default. Projects like Aztec are launching, but they’re standalone rollups. How do we balance these factors—meeting user needs while gaining state acceptance? What’s the realistic roadmap for privacy?
Vitalik:
I think this splits into two parts. First, how to turn privacy from a niche feature into a default UX; second, how to make governments and regulators comfortable with it.
My hesitation about implementing privacy directly on Layer 1 isn’t because I think it’s fundamentally wrong, but because I don’t believe the tech is mature enough. We don’t yet know which privacy tech is optimal. Implementing a specific tech on L1 could lead to bad outcomes. This is a challenge in any EIP, but especially with privacy due to data sensitivity—you can’t easily swap tree structures, making upgrades harder.
David:
Does this mean you think implementing privacy on Ethereum Layer 1 is the fundamentally right direction?
Vitalik:
Long-term, I’m open to it. It’s a balance between futurecompatibility and security. Remember, bugs in L1 privacy tools could be catastrophic—someone could mint unlimited tokens undetected. I believe as tech matures, we’ll get there.
In a recent blog post, I noted that code bugs are actually decreasing. We’ll reach a point where code is highly trusted—something security researchers couldn’t imagine 20 years ago. I think AI will accelerate this. Until then, projects like Zcash have bravely pioneered, but they face extremely high technical bars.
The question is, can we make privacy as close to default as possible without putting it directly on Layer 1? My medium-term goal is making privacy features default in wallets. A major mistake in the ecosystem is creating the concept of “privacy wallets.” We shouldn’t have “privacy wallets”—privacy should be a feature of all wallets. Privacy features should seamlessly integrate into existing wallets.
Ryan:
So, if in my existing crypto wallet browser extension, I could choose to send a normal transaction or a private one—would that default setting work?
Vitalik:
Exactly. You should have a private balance and a private send button—features built into Metamask or other wallets. The Ethereum Foundation has already started related work—hopefully we’ll see progress in the coming months.
Privacy Resistant to State Intervention
Ryan:
On the other side of privacy tech—how do we gain state acceptance? Clearly, projects like Ethereum want to push privacy, open source, and decentralization. But these ideals may clash with some governments and cultures. We hope to advance social culture through tech, while gaining mainstream acceptance instead of global bans. Is there a balance—addressing state concerns about crime and money laundering while preserving cypherpunk values? Or are these fundamentally incompatible?
Vitalik:
First, I think privacy pools have made significant progress. Protocols like Railgun are now usable—we see success cases proving they can prevent stolen funds from major DeFi contracts from entering privacy pools. This shows privacy pool tech is maturing and passing real-world tests.
Currently, most illicit funds flowing through privacy protocols come from DeFi hacks or individual account thefts. In these cases, privacy protocols can use blacklists to restrict flows. If your funds are stolen, you can mark them suspicious via API; if a DeFi project is hacked, similar measures apply. So this mechanism is already key to advancing privacy tech.
Still, we must acknowledge that today’s blockchain ecosystem can’t fully resist malicious actors. Bad actors still have many ways to move and hide funds. Yet compared to traditional finance, transparent public blockchains offer some traceability. If your funds go missing, you can track their path. I believe privacy tech can boost resistance to bad actors while offering more privacy to ordinary users—a balance we can achieve.
On the enforcement side, we need strong arguments for privacy tech. For example, telecom surveillance data has recently leaked due to hacks. This data was collected per the Communications Assistance for Law Enforcement Act (CALEA) and given to governments, but leaks could threaten national security—other nations or groups might exploit it. Such incidents happen globally.
A bloated centralized data collection system is fragile. If these databases are hacked, data meant to aid national security could actually destabilize it. I think this argument needs stronger advocacy. Two lawmakers told the public that actively reducing data collection is actually safer. That’s the direction we should move—privacy-preserving finance is part of it.
Cypherpunk vs Mainstream Culture
Ryan:
I feel you want to maintain a firm stance on privacy in crypto and Ethereum. This ties into broader cultural issues. Over the past decade, cypherpunk ideals were attempted but fell short. Now Ethereum stands at a historic moment—gaining mainstream acceptance. Robinhood is moving into Layer 2, JP Morgan talks about on-chain business, Coinbase grows stronger. Traditional finance and mainstream culture are entering Ethereum.
But this brings challenges. How will people who don’t value cypherpunk ideals affect Ethereum? You mentioned Ethereum’s pluralism—we welcome more people, but they may not cherish our core values. So where do we draw the line? Projects like Rye stablecoin may align better with cypherpunk ideals but failed. Meanwhile, Circle and Tether, while less ideal, have helped people in emerging nations. How do we balance cypherpunk values with real-world needs? When should product-market fit and actual usage take priority?
Vitalik:
I think privacy and cypherpunk values must be prioritized in certain areas—first, the base protocol. A decentralized backend can support both centralized and decentralized apps, but a centralized backend can only support centralized apps. If the blockchain itself doesn’t support privacy or trustless interactions, there’s no space for privacy-friendly apps.
For example, I’ve pushed for account abstraction and improving EIP 7770—because otherwise, many smart wallet use cases—like multisig, quantum resistance, and privacy protocols—might rely on intermediary ecosystems. The problem with intermediaries is their limitations—if they fail, users suffer. So we must ensure Ethereum interactions are at least privacy-friendly and don’t depend on centralized intermediaries for basic operations.
Second, we must ensure privacy-preserving, trustless mechanisms are at least feasible. This spans protocols, wallets, and apps. Most users may not choose this path, but we can’t let infrastructure gaps or hostile top-layer protocols and standards shut it out.
Take email: theoretically, it’s an open protocol—anyone can run a server. But due to spam, big providers blacklist small servers. So email relies heavily on centralized permissioning, not openness. We need to better protect email’s openness.
Similarly, for new Ethereum standards, we must ensure privacy-friendly trustless operations are feasible, support self-sovereignty, and avoid unnecessary performance loss. Even if not everyone chooses it, that’s fine. People can still hold ETH on Coinbase—but the option must exist, and protocols/standards must consider these needs.
David:
Before diving into Ethereum specifics, I’d like to ask a macro question for context. Future trends are emerging—like AI shaping the future. We can debate possibilities—geopolitical tensions increasing, the world fragmenting, alongside exciting advances like gene editing. Vitalik, what role do you see Ethereum playing in these future trends?
Vitalik:
I’m glad we agree on gene editing and AI. Ethereum’s role has two parts.
First, it provides tools protecting people’s freedom, autonomy, and organizational capacity—tools independent of any individual, company, or nation. In an increasingly fragmented world, this is crucial. Fifteen years ago, many trusted Facebook. Now we broadly agree only blockchains are trustworthy. So the market for trust tech is ready—many want to pay for it, care about social issues, and seek these techs as solutions. This lays a solid foundation.
Second, building a global community. Ethereum attracts those who care aboutdecentralized finance, innovative organizational forms, privacy, and democratic governance. This intellectual magnetism makes Ethereum a unique community asset. Even in a hypothetical future—say, tomorrow we solve NP problems, making all blockchains obsolete—Ethereum’s community itself retains irreplaceable value.
World Ledger
Ryan:
In Ethereum’s early days, people often called it a “world computer.” But that term faded. Earlier this year, you described Ethereum as a “world ledger”—I find this more accurate and concrete. Though still abstract for some, it suggests Ethereum can be a global tool for recording key info like property.
I’ve been pondering: what is Ethereum? I lean toward calling it a decentralized property rights system. While that may confuse laypeople, do you think “world ledger” accurately captures Ethereum’s core mission?
Vitalik:
I think “world ledger” is indeed more accurate. Compared to “world computer,” “ledger” is more concrete and better conveys Ethereum’s core value. “World computer” is too broad—computers can generate cat pictures, translate text, or make videos—none suitable for Ethereum, especially on L1.
“Ledger” focuses on economic value—not just DeFi, but also ENS (Ethereum Name Service). So I say Ethereum L1 is a global ledger—this clarifies its role and explains the L1-L2 relationship better. The goal is helping people understand Ethereum’s place. If people accept this definition, it’s working.
David:
If Ethereum is the world ledger, what is ETH?
Vitalik:
That’s an interesting question. If we compare the ledger to a book, ETH is like the “ink” used to write it. We can extend this: ink represents L2 functionality. If the ink has a TM (trademark), it’s a branded L2; without TM, it’s unbranded. This metaphor may help clarify ETH’s role.
Ethereum in 2024: Outlook
David:
I think your points are valid. Next, let’s discuss Ethereum’s state in 2024. Personally, I think 2024 was tough for Ethereum. ETH’s price stayed low—impacting the whole ecosystem and causing internal divisions. Your thoughts? Was 2024 truly difficult for Ethereum? How would you explain it? If telling the story of Ethereum in 2024, how would you frame it?
Vitalik:
I think ETH’s sustained low price is indeed a core reason behind many issues. For many in the community, another key factor is that once-popular topics and projects lost momentum, while new alternatives haven’t emerged. For example, some had high hopes for DAOs, but when DAOs stalled, they felt lost—unsure what to focus on next.
Also, the NFT market cooled in 2024, while memes exploded—but meme growth happened mostly on Solana, not Ethereum. Ethereum’s core values promote freedom and openness, but meme popularity hasn’t significantly advanced these goals—leaving many confused.
Another major challenge is the L1-L2 relationship. Some in the community feel L1-L2 collaboration is broken—especially regarding incentives. When people realize interests may diverge, internal divisions deepen.
2024 resulted from multiple factors. On one hand, ETH’s price drop pressured the ecosystem; on the other, once-hot projects and topics faded while new ones haven’t successfully replaced them. I think the path forward must involve new projects deliveringtangible benefits with broad support. Actually, I feel by 2025 we’re seeing positive changes—perhaps these issues are gradually resolving. That’s part of why we now feel more hopeful.
Changes at the Ethereum Foundation
David:
Currently, the Ethereum community is still debating 2024’s issues. Some may be narrative psyops, others real challenges needing solutions. A core topic is that Ethereum’s “rollup-centric roadmap” seems off-track. Chain fragmentation is severe—each chain and L2 feels isolated, and L1 isn’t truly scaling. This roadmap seems misaligned with Ethereum’s goal of being a unified global computer.
Meanwhile, the Ethereum Foundation has adjusted. Can you explain why these changes occurred? What does the foundation need to improve? What’s its current state?
Vitalik:
I think these changes were brewing—they just needed triggers.
The leadership shift at the Ethereum Foundation is a classic example. Aya now chairs the board—she seems more at ease in this role while still engaging in specific projects, like promoting Ethereum in Bhutan. She’s also deeply interested in Russell and financial inclusion, dedicating more time. This shows the executive director role isn’t suited for one person long-term.
My tenure was relatively long, so leadership changes include my own role adjustments. The foundation also added new voices and priorities—Tom and Shelley, each with tech expertise.
Meanwhile, many new leaders emerged across departments and projects. Work on scaling and UX is being restructured. One area I’m focused on is better organizing censorship resistance and privacy efforts—shifting from theory to deploying privacy tech in production. Many new, exciting projects have launched. These changes would’ve happened eventually, but key events accelerated them.
Regarding L1-L2, I don’t think it’s directly tied to foundation changes. Clearly, increasing L1 gas limits is a key foundation agenda, and we’re also improving L2 interoperability.
Work started mid-2024 but accelerated due to early-2025 events. L2 interoperability improvements continue—many L2s completed phase one. Next, focus is shortening withdrawal times. Goal: reduce from one hour to minutes—or even 12 seconds. Depends on gas users pay. If withdrawals take hours or weeks, native deposits/withdrawals become inefficient.
So long-term, assets issued via L1-trust models may lose advantage. More competitive are bridges relying on custom mint/burn—ultimately concentrating power in multisig control. To make L1-based methods competitive, we must drastically shorten withdrawal times.
If withdrawal time drops to one hour, liquidity becomes cheaper—possibly free. At 12 seconds, more users might use L1 for deposits/withdrawals. This requires ecosystem-wide effort—foundation teams, L2 projects, zkEVM devs. Overall, it’s a complex multi-party collaboration needing time and resources.
I’m optimistic about progress. We need a strong L1 and clear L1-L2 relationships. Though all technical steps take time, we’re moving in the right direction.
Economic Collaboration
Ryan:
I remain concerned about the rollup roadmap. Though we’ve seen success in rollup deployment—like RobinHood’s recent push—I worry we’re not paying enough attention to different stages of user property rights on Ethereum L2.
Phase one progressed, and I think phase two will too. But we’ve underemphasized coordination between L1 and L2. While some UX and standard issues can be fixed tactically, I’m more worried about the lack of overall collaboration. If we reach a state, as you mentioned in 2021, where Ethereum is decentralized, L1 is decentralized, but one large rollup chain centralizes all execution states. Then I worry L2 might dominate rules, possibly even脱离 Ethereum. This seems economically unbalanced. Relationships between chains and brands also feel looser than past Ethereum L1 ties. Your thoughts on economic collaboration? Can we fix it? Or will it take a long time?
Vitalik:
I agree. I think economic collaboration can be viewed two ways. One is L2 fee structure. Currently, L2 base fees are too low—I think moderately higher fees could fix many issues. But I’m unsure fees are the key variable. I think network effects matter more.
Another key concern is withdrawal time. If all assets are issued on L2 and users transfer via mint/burn bridges, L1 engagement drops sharply. Keeping L1 central is vital—even if most activity is on L2, assets should still be issued on L1. This is safer trust-wise and solves L2 governance/upgrades. If assets are issued on L2, users must trust that L2 regardless of transfers. But if issued on L1, users can rely on L1’s finality via native withdrawals. This is more secure and enables freer movement between L2s—supporting cross-L2 apps.
Therefore, encouraging asset issuance on L1 and making it economically viable is crucial to maintaining L1’s centrality. I think we should strive for this.
Also, raising base fees to a reasonable level helps. Plus, we should explore synchronous composable blocks—letting L2 interact with L1 in real time. This makes L2 value derive more from L1 collaboration—not just being a bridged standalone chain.
Scaling L1
Ryan:
David and I mainly focus on Ethereum’s social layer. We believe that to achieve strong impact quietly and boost L1’s soft power, we need a stronger L1. This isn’t just about boosting TPS—it’s about making L1 the core of DeFi, a liquidity hub, and primary asset issuance platform. The stronger L1 is, the more effectively it can attract and guide L2s, providing liquidity—why they stay tightly linked to Ethereum. We’re excited about initiatives to boost L1 performance.
In recent discussions, new trends seem to emerge—especially under the new Ethereum Foundation (EF) leadership—making L1 performance enhancement the top priority. From your view, why is boosting L1 performance so important?
Vitalik:
The key question is always how to scale L1 securely. Security means not destabilizing the network, not fully centralizing node operations, not breaking the staking ecosystem. The tech we now have—like ZKVMs—didn’t exist a few years ago. Now it’s nearly production-ready; a year ago, impossible. So many are pushing this tech hard.
We plan to raise block gas limits 3–5x. If this eliminates the bottom 10% of solo stakers due to resource limits, we don’t ignore them—we hope they use ZKVMs to verify the chain instead of manually re-executing everything. This is safe—even if 10% of nodes use ZKVMs, as long as it stays below one-third.
As ZK tech matures, we can strategically apply it across domains. This marks L1’s first phase. As safety improves, L1 enters a second phase.
Other tech can also boost L1 performance. For example, historical storage has long consumed massive node space. We recently implemented basic historical data cleanup—removing pre-merge data—saving hundreds of GB per node. Going forward, we’ll optimize further—aiming to clean history after each upgrade, with long-term goal of data expiring after 36 days. We’ll need P2P distributed storage to ensure full chain accessibility and verifiability.
Another promising tech is gas repricing. Block Access OS is also key—it lets every node (except block proposers) execute blocks with maximum parallelism. Proposers must execute sequentially, but their hints allow others to validate and re-execute in parallel. This safely runs Ethereum at higher gas throughput.
So now we have many well-optimized tech options—more tools than ever to balance scalability and security.
Ryan:
Combining these technologies, do you think we can achieve Don Krad’s goal while keeping Ethereum decentralized? He believes Ethereum’s transaction volume can grow from today’s ~20 TPS to 10,000 TPS in 3–5 years. Do you think this is realistic?
Vitalik:
My confidence in extreme targets has waned, but the issue is mainly ultra-short block times, not ultra-high TPS. If I had to choose, I’d prefer 10,000 TPS with 12-second blocks over 1-second blocks. In extreme cases, you hit basic decentralization issues like light speed. So personally, I advocate caution on extreme metrics—but I do believe there’s room for optimization.
From a decentralization perspective, we’re in a position hard to judge, but actually I hope decentralization improves in many key aspects. Let me illustrate. People often complain few actually run nodes, relying on RPC services. Now we finally have a reliable roadmap out of this. Part of it is Helios—light clients running in wallets, with improving efficiency.
Another reason is that cypherpunks typically value privacy from personal node operation. I didn’t grasp this until talking with some. Running your own node offers strong privacy protection. For example, if you use Ethereum privacy protocols, operate across multiple accounts, and care deeply about privacy—but still query each address balance via Infura—Infura knows all your links. Running your own node avoids this—you download blockchain data locally, all queries are local, no one sees your reads.
So how to achieve both? Soon we can. Two paths. Path one: improve efficiency from the current Ethereum node concept. Use aggressive data expiry, don’t store all data. You don’t even need branch storage—just a state table, needing only 80 GB.
With block-level access lists and zkVMs verifying their correctness, you can keep state updated with minimal computation—maintaining just an 80 GB local DB. 80 GB is tiny—equivalent to 3–4 LLM models. Almost every device has 80 GB—my phone does.
But if we expand L1 moderately—say 30x—80 GB becomes 2.4 TB. 2.4 TB returns to today’s full node storage needs. We’ve already granted ourselves 30x expansion. For further scaling, we can adopt partial state nodes—storing only state related to top 100 apps plus all EOAs and contract data. This greatly reduces storage.
Path two: start from browser wallets and add more safeguards. First is Helios—light clients verifying the chain. Second, use near-term tech like TEEs and Oram. Long-term, use PIR (private information retrieval)—cryptographically ensuring servers don’t know request content. Servers respond without knowing what they responded to. Standardizing this gives strong privacy.
For example, you could have light clients offering full-node security and privacy without running a full node. We have multiple paths to deliver decentralization properties unimaginable in 2017.
Even in early stages—before L1 scales significantly—I think we’re achieving greater scale, decentralization, privacy, and censorship resistance. Still, concerns remain—ensuring PoS and block building stay decentralized is key. Having many high-quality research teams focusing on different areas and methods will be very valuable.
Overall, I think we’re making great progress increasing both decentralization and scale—we just need to keep pushing.
Ethereum’s Barbell Strategy
David:
“If Ethereum L1 directly entered thehigh-frequency trading (HFT) game, it would fundamentally damage its core values, because over-pursuing HFT would divert Ethereum from its original purpose.” I don’t fully grasp this conclusion. Can you explain why participating in HFT harms L1’s core values? How does Ethereum’s L1 scaling strategy address this?
Vitalik:
The core reason is, if we over-optimize one aspect—likelow latency forHFT—we sacrifice other critical attributes, potentially unbalancing the entire ecosystem. This resembles AI safety debates: unchecked pursuit of certain technologies can bring unintended negative outcomes.
InHFT,low latency is everything. To minimize latency, ecosystem participants are incentivized to endlessly optimize setups—with no natural stopping point. Eventually, this could erode Ethereum’s global decentralization. For example, setting block time to 1 second forces propagation and confirmation under 500ms each. This pushes extreme P2P communication optimizations and strong “co-location incentives.”
“Co-location incentives” mean participants place servers as close as possible to block proposers to reduce latency—significantly boosting speed and competitiveness. As a DeFi participant, you want to send trades fast to capture market moves; as a dev, shaving 5ms might let you send 1% more transactions per block, increasing profits.
The reality ofHFT systems is all participants cluster servers together chasing minimal latency. So I think Ethereum’s “dual strategy” is sound: L2 handles centralized tasks like HFT; L1 focuses on security and censorship resistance. This way, L2 isolates L1 from co-location incentives, avoiding excessive centralization.
L2’s independent ordering is one of its advantages. That’s why my interest in Base Layer has waned lately, while focus on L2 grew. L2 can leverage centralization for efficiency while staying decentralized in governance—a powerful balance. Thus, L2 is well-suited forHFT needs.
Looking ahead, AI thinks 1,000x faster than humans. From an AI’s view, light speed is only 300 km/s. With AI, a global financial ledger may no longer make sense. We’ll need city-level ledgers—where L2s are a natural fit.
If L1 heads this way, centralized incentives accumulate. Committing to compete in HFT means extreme optimization in environments where others don’t value decentralization.
So I believe the “dual strategy” is correct. L1 does need improvement—especially block time. Bitcoin’s 10-minute block time is clearly too long. If Bitcoin used 20-second blocks, its trajectory might’ve been vastly different.
L1 should have moderately low latency for average users; for ultra
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
深潮TechFlow
