Quick Read: Key Points from Ethereum Foundation AMA — L1 Revenue and Value Accumulation, Pectra Upgrade, L2, and More
TechFlow Selected TechFlow Selected
Quick Read: Key Points from Ethereum Foundation AMA — L1 Revenue and Value Accumulation, Pectra Upgrade, L2, and More
L1 revenue and value accrual, L2, blob fees, L1 gas limit target, risk of big corporations taking over Ethereum, Pectra upgrade...
Navigating Web3 tides with focused insightsCompiled & edited by: KarenZ, Foresight News
On February 25, the Ethereum Foundation research team hosted its 13th AMA on Reddit. Foresight News reviewed over 300 comments and compiled key insights from Vitalik Buterin and members of the Ethereum Foundation research team, covering topics such as L1 revenue and value accrual, L2s, blob fees, L1 gas limit targets, risks of large corporations taking control of Ethereum, Pectra upgrade progress, and other future plans.
Fees
Question: The blob fee model seems somewhat unsatisfactory—overly simplistic in a way—by setting the minimum fee at the smallest unit of ether that exists in the protocol (1 Wei). Given how EIP-1559's pricing mechanism works, during aggressive expansion of blob usage, we might see long periods with no blob fees at all. This doesn't seem ideal—we should incentivize blob usage but not make it free on the network. With this in mind, are there plans to restructure the blob fee model? If so, how? What alternative fee mechanisms or adjustments are being considered?
Vitalik Buterin: I do believe we should keep the protocol simple, avoid over-adapting to short-term conditions, and align logic between execution gas and blobs in the gas market. Ethereum Improvement Proposal 7706 (EIP-7706) focuses on this as one of its two main goals (the other being adding a separate gas dimension for calldata).
Ansgar Dietrichs: Max Resnick proposed a potential solution in EIP-7762. The proposal suggests setting the minimum fee low enough to effectively be zero cost during non-congested times, yet high enough to ramp up fees more quickly when demand increases. This proposal came relatively late in the Pectra hard fork development cycle, and implementing it could risk delaying the fork. We brought this issue to RollCall #9 to assess whether the problem is severe enough to justify such a delay; see https://github.com/ethereum/pm/issues/1172. Feedback received indicated that L2 teams no longer view this as an urgent issue. Based on this, we decided to maintain the current model in the Pectra hard fork. However, if there's sufficient ecosystem demand, this remains a viable feature option for future hard forks.
Dankrad Feist: Concerns about blob fees being too low have been greatly exaggerated and are short-sighted. That said, in the short term, I do think setting a higher minimum price for blobs would be better.
Justin Drake: Yes, EIP-7762 could increase MIN_BASE_FEE_PER_BLOB_GAS from 1 WEI to something higher, like 2 ** 25 WEI.
Question: What are the Ethereum Foundation’s plans for improving scalability and reducing mainnet transaction fees in the coming years?
Vitalik Buterin:
-
Scaling L2: More blobs (e.g., PeerDAS in Fusaka).
-
Continuing to improve interoperability and cross-L2 user experience (e.g., see the recent Open Intents framework).
-
Moderately increasing the L1 gas limit: See this for rationale.
Ethereum Value Accrual and Token Price
Question: L2 scaling has led to significant loss in value accrual for L1, which also affects ETH. Besides “L2s will eventually burn more ETH and conduct more transactions,” what plans do you have to address this issue?
Justin Drake: Blockchains—whether L1 or L2—typically have several sources of income. First is congestion fees, i.e., "base fees." Second is competitive fees, i.e., MEV (Maximal Extractable Value).
Let’s first discuss competitive fees. In my view, as modern application and wallet designs evolve, MEV will increasingly flow upstream and be recaptured by applications, wallets, and/or users. Ultimately, nearly all MEV will be recaptured by entities closer to where traffic originates, while downstream infrastructure like L1 and L2 will capture only a small fraction from competitive fees. In other words, in the long run, L1 and L2 chasing MEV may be futile.
What about congestion fees? For Ethereum L1, the historical bottleneck was EVM execution. Consensus participant constraints such as disk I/O and state growth were key drivers behind keeping the execution gas limit small. With modern blockchain designs using SNARKs or fraud proof games for scaling, we’re moving into an era of post-execution scarcity. The bottleneck then shifts to data availability (DA), which is inherently scarce because Ethereum validators operate on limited home internet connections, and DAS can only offer roughly linear ~100x scalability improvements, unlike fraud proofs and SNARKs which provide near-infinite scalability gains.
Therefore, let’s dive deeper into DA economics, which I believe is the only sustainable revenue source for L1. EIP-4844 significantly increased DA supply via blobs, effective less than a year ago. The chart titled “Average Number of Blobs per Block” on the dashboard clearly shows growing blob demand over time (driven largely by induced demand), gradually rising from 1 blob per block to 2, then 3. We are now saturating blob supply, but still in the early stages of blob price discovery, with low-value “junk” transactions being economically displaced by denser, more valuable ones.
If DA supply remains unchanged for months, I expect hundreds of ETH per day to be burned due to DA. However, currently Ethereum L1 is in “growth mode.” The upcoming Pectra hard fork (in a few months) will increase the target number of blobs per block from 3 to 6. This surge in DA supply should suppress the blob fee market, requiring months for demand to catch up again. As full danksharding rolls out over the next few years, there will be a cat-and-mouse game between DA supply and demand.
What will long-term equilibrium look like? My argument since the 2022 Devcon talk “Ultra Sound Money” hasn’t changed. In the long run, I expect DA demand to exceed supply. Indeed, supply is fundamentally constrained by consensus participants running on home internet connections. I believe DA throughput equivalent to about 100 home internet connections won’t suffice for global demand—especially given humans’ endless creativity in consuming more bandwidth. In about 10 years, I expect Ethereum to reach 10 million TPS (roughly 100 transactions per person per day), generating $1 billion in daily revenue even at $0.001 per transaction.
Of course, DA revenue is just one part of ETH’s long-term value accrual. Two other important factors are issuance and monetary premium.
Dankrad Feist: All blockchains face value accrual challenges, and there’s no perfect solution. Execution layers perform slightly better than data layers because they can extract priority fees reflecting transaction urgency, whereas data layers charge fixed fees. My answer to value accrual starts with creating value. While doing so, we should maximize opportunities that could lead to future revenue. This means maximizing the value of Ethereum’s data layer, enhancing Ethereum’s overall value so alternative data availability (alt DA) isn’t needed; scaling L1 so high-value applications can truly run on it; and encouraging projects like EigenLayer that expand Ethereum’s use as (non-financial) collateral.
Question: If the price of Ethereum drops below a certain level, would the economic security of ETH be threatened?
Justin Drake: High economic security is crucial if we want Ethereum to be truly resistant to attacks—including those from nation-states. Currently, Ethereum has approximately $80 billion in economic security (slashable), the largest among all blockchains (33,644,183 ETH staked, current ETH price $2,385). By comparison, Bitcoin’s economic security is around $10 billion (non-slashable).
Question: What is the ticker?
Justin Drake: For me, it’s ETH. I also hold some BTC, mostly for sentimental reasons, as a collectible.
L2
Question: Regarding L2 interoperability, many websites (e.g., Aave, Uniswap) and wallets (e.g., MetaMask, Trust Wallet) now have increasingly long dropdown menus to select different L2 networks, leading to poor user experience. When can we expect these dropdown menus to disappear completely?
Vitalik Buterin: I hope chain-specific addresses can reduce the need for such dropdowns in many scenarios. You could paste an address like eth:ink:0x12345...67890, and the app would immediately know you want to interact with Ink and handle the backend accordingly. In many cases, this is more of an application-specific challenge—finding best practices to hide complexity from users. Another long-term possibility is better cross-L2 interoperability, enabling more DeFi apps to run on just one major L2.
Question: Given community sentiment around Ethereum, do you still firmly believe focusing on L2 solutions is the winning approach? If you could go back in time, would you change anything?
Ansgar Dietrichs: In the long run, rollups remain the only principled way to scale blockchains to the size required for a global economic base layer. Looking back, I think we didn’t invest enough effort in the path toward this end goal and intermediate user experience. Even in a rollup-centric world, L1 still needs significant scaling (as Vitalik recently outlined). We should recognize that advancing both L2 efforts and parallel L1 scaling paths delivers better value to users during the transition period.
My view is that Ethereum didn’t face strong competition for a long time, so it became somewhat complacent. The fiercer competition we now see highlights some misjudgments and forces us to deliver a better overall “product”—not just theoretically correct first-principles solutions. But yes, to reiterate, some form of rollup is essential for achieving the “endgame of scaling.” Specific architectures are still evolving—for example, Justin’s recent native rollup exploration shows methods are still changing—but the overall direction is clearly correct.
Dankrad Feist: I disagree with parts of this answer. If you define rollups merely as “scaling verification of DA and execution,” how are they different from execution sharding? In reality, we think of rollups more as “white label Ethereum.” Fairly speaking, this model unleashed massive energy and capital. Had we focused solely on execution sharding in 2020, we wouldn’t have made such progress in zkEVM and interoperability research today. Technically, we can now achieve whatever we want—a highly scalable L1, a more scalable sharded blockchain, or a rollup base layer. In my opinion, Ethereum’s best path is combining the first and third options.
Future Plans and Discussions
Question: On timelines of short-term (less than 1 year), 1–3 years, and beyond 4 years, what types of applications will be designed for Ethereum?
Ansgar Dietrichs: This is a broad question, so I’ll give a (very) partial answer focused on broader trends.
I firmly believe we’re at a pivotal turning point in crypto history. We’re emerging from a long “sandbox” phase where crypto primarily focused inward—building internal tools, creating infrastructure, developing foundational modules like DeFi—with limited real-world integration. All of this was important and valuable, but had limited real-world impact.
The current moment aligns with both technical maturity (though work remains, we broadly understand how to build infrastructure supporting billions of users) and positive regulatory shifts in the largest market (the US). Taken together, I believe it’s time for Ethereum and crypto as a whole to move beyond the sandbox.
This shift will require fundamental changes across the ecosystem. The best articulation of this challenge I’ve seen is DC Posch’s “Real World Ethereum” vision: https://daimo.com/blog/real-world-ethereum. The core theme is building real products for real people in the real world, using crypto as an enabler—not the selling point itself. Importantly, all of this still preserves our core crypto values.
Currently, the main type of real-world product is stablecoins (which started earlier due to fewer regulatory hurdles), along with a few smaller “real-world impact” success stories like Polymarket. In the short term, I expect stablecoins to leverage their first-mover advantage to grow further in scale and importance.
In the medium term, I anticipate real-world activities becoming more diverse: other real-world assets (such as stocks, bonds, and anything representable on-chain). Beyond assets, I predict we’ll see many new types of activities and products (e.g., mapping business processes on-chain, governance, novel mechanisms like prediction markets).
All of this will take time, but efforts invested here will pay off long-term. Over-focusing on continuing “sandbox” activities (e.g., meme coins) might show more appeal in the short term, but risks being left behind as Real World Ethereum takes off.
Carl Beekhuizen: Overall, we focus on scaling the entire tech stack rather than designing for specific applications. The overarching theme is scalability: how to build the most powerful platform possible while maintaining decentralization and censorship resistance.
In the short term (<1 year), the main focus is launching PeerDAS, which will allow us to significantly increase the number of blobs per block. We’re also improving the EVM: we hope to launch EOF soon. Significant research is going into statelessness, EOF, gas repricing, ZK-EVM (zero-knowledge EVM), and related areas.
Over the next 1–3 years, we’ll further scale blob throughput and launch some of the research projects listed above, including advancing zkEVM initiatives such as ethproofs.org.
Looking ahead to 4+ years, our vision includes adding a suite of extensions to the EVM (which L2s will adopt for acceleration), significantly increasing blob throughput, improving censorship resistance (e.g., via FOCIL), and accelerating everything further through various ZK (zero-knowledge) techniques.
Question: There’s a view that one day the Ethereum mainnet should become immutable, with innovation happening at the L2 layer. Yet we continuously see new research (like execution tickets, APS, one-time signatures, etc.), and the Ethereum Foundation is actively pushing these—which is great. The landscape keeps changing, and from my experience, digital products are “never finished.” In other words, after Vitalik’s roadmap / beacon chain implementation, how likely is it that we’ll still need adjustments?
Vitalik Buterin: Ideally, we could separate parts that can become immutable from those needing continuous evolution. We’re already doing this to some extent through the separation of execution and consensus (with bolder consensus upgrades, including Justin Drake’s recent idea of a full beacon chain overhaul). I expect these specs to continue evolving. Additionally, I believe we’re seeing “light at the end of the tunnel” for many technical problems, as the pace of research has indeed slowed compared to about five years ago, with recent focus shifting more toward incremental improvements.
Question: Vitalik commented in a recent article on Verge: We’ll soon face a decision point on which of three options to pursue: (i) Verkle trees, (ii) STARK-friendly hash functions, (iii) conservative hash functions. Has a path been decided?
Vitalik Buterin: It’s still under discussion. Personally, I sense the sentiment has slightly leaned toward (ii) over the past few months, but no decision has been made. I also think it’s worth considering these options within the context of the broader roadmap they’d be part of. Specifically, the most realistic options to me seem to be:
Option A:
2025: Pectra, possibly EOF
2026: Verkle
2027: L1 execution optimizations (e.g., delayed execution, multi-dimensional gas, repricing)
Option B:
2025: Pectra, possibly EOF
2026: L1 execution optimizations (e.g., delayed execution, multi-dimensional gas, repricing)
2027: Initial rollout of Poseidon
2028: Gradual adoption of more stateless clients over time.
Option B is also compatible with conservative hash functions; however, in that case, I’d still favor a gradual rollout, as even with lower risk than Poseidon, proof systems carry higher initial risk.
Justin Drake: As Vitalik said, it’s still under discussion. That said, long-term fundamentals clearly point to (ii). Indeed, (i) lacks post-quantum security, while (iii) is inefficient.
Question: Any recent progress on VDFs?
Dmitry Khovratovich: A 2024 paper revealed a potential attack on the candidate VDF MinRoot, showing computation can be accelerated on multi-core machines, breaking its sequentiality. Currently, there’s a lack of efficient and secure VDF schemes (efficiency meaning computable on small hardware, security meaning non-acceleratable), and no reliable VDF candidates. Therefore, VDF research and applications have been temporarily put on hold.
Question: Is there willingness to scale Ethereum 100x next year? How open are you to simple parameter tweaks in the protocol? For example, shortening block time by 3x, doubling the block limit, raising gas targets, increasing blob count, etc.?
Francesco D'Amato: Scaling Ethereum overall by 100x isn’t realistic, but achieving a 100x increase in blob throughput compared to pre-4844 levels is possible. EIP-4844 already delivered ~3x scaling, Pectra is expected to add another 2x, and Fusaka aims for 4–8x. That leaves us needing another 2–4x. I believe we definitely have ways to achieve that.
Question: What features will the Fusaka & Glamsterdam upgrades include?
Barnabé Monnot: Fusaka appears to focus primarily on PeerDAS, which is critical for scaling L2s, and few want to delay Fusaka for other features. Personally, I’d love to see FOCIL and Orbit in Glamsterdam, paving the way toward SSF (Single Slot Finality). These focus more on the consensus layer (CL) and data availability (DA), but Glamsterdam should also push L1 scaling on the execution layer (EL), with ongoing discussions about which feature sets are most suitable.
Question: Can an EIP “force” L2s to adopt Stage 1 (or even Stage 2) decentralization (given their slow progress)?
Vitalik Buterin: Native rollups (e.g., EXECUTE precompiles) do this to some extent. L2s remain free to ignore the feature, write their own code, or even add backdoors, but they’ll have access to a simple, highly secure proof system built directly into L1, so L2s seeking EVM compatibility will likely choose this option.
Question: After Fusaka/Glamsterdam, which research projects might be ready for development upgrades?
Toni Wahrstätter: PeerDAS is moving forward aggressively, along with proposals like EOF, FOCIL, ePBS, SECP256r1 precompile, and delayed execution. PeerDAS is now ready for inclusion in the Fusaka upgrade, and there’s broad consensus on its urgency. The other proposals mentioned could be candidates for Glamsterdam, but which EIPs will be included hasn’t been finalized.
Question: Vitalik wrote about proposed measures in case of a quantum emergency. How would we determine that we’re in a quantum emergency?
Vitalik Buterin: In practice, combining media reports, expert opinions, and Polymarket predictions on when a “true” quantum computer (capable of breaking 256-bit elliptic curve cryptography) will emerge. If the timeline is within 1–2 years, that’s definitely an emergency; if around 2 years, not an emergency but urgent enough to drop other roadmap priorities and integrate all quantum-resistant technologies into live protocols immediately.
Question: What is the L1 gas limit target for 2025?
Toni Wahrstätter: There are many differing views on gas limits, but it ultimately comes down to a key question: Should we scale Ethereum L1 by increasing the gas limit, or should we focus on L2 and enable more blobs using advanced techniques like DAS (data availability sampling)?
Vitalik recently published a blog post discussing moderate L1 scaling, listing reasons why increasing the gas limit could make sense. However, increasing the gas limit involves trade-offs: higher hardware requirements; faster state and history growth; increased bandwidth demands.
On the other hand, Ethereum’s rollup-centric scaling vision aims for greater scalability without increasing node hardware requirements. Technologies like PeerDAS (short-term) and full DAS (medium-to-long term) are expected to unlock significant scaling potential while keeping resource demands reasonable.
That said, I wouldn’t be surprised if after the Pectra hard fork in April, validators push the gas limit up to 60 million. But in the long run, the primary scaling focus will likely center on DAS-based solutions rather than simply increasing the gas limit.
Question: If the Ethereum beam client experiment (or whatever it ends up being called) succeeds, and in 2–3 years we have several working implementations of Ethereum beam clients, would we need a phase where current PoS and beam PoS run in parallel, both earning staking rewards—similar to the PoW + PoS parallel phase before the PoS transition?
Vitalik Buterin: I think we can do an immediate upgrade. The reason two chains ran in parallel during the merge was:
-
PoS as a whole was untested, and we needed time for the entire PoS ecosystem to launch and run long enough to gain confidence before switching.
-
PoW could undergo reorgs, and the switch mechanism needed to be robust against them.
But PoS provides finality, and most infrastructure (e.g., staking) will carry over. So we can do one big hard fork to switch validation rules from the beacon chain to the new design. Perhaps at the exact moment of transition, economic finality guarantees aren’t fully satisfied, but to me, that’s a small and acceptable cost.
Question: The Ethereum Foundation launched a $2 million academic grant program for 2025. Which research areas are prioritized? How does the Foundation plan to integrate academic research outcomes into the broader Ethereum development roadmap?
Fredrik Svantes: Here’s a wishlist: https://www.notion.so/efdn/17bd9895554180f9a9c1e98d1eee7aec.
Research directions of interest to the protocol security team include:
-
P2P security: Many vulnerabilities we’ve found relate to denial-of-service vectors at the network layer (e.g., libp2p or devp2p), so improving security here would be highly valuable.
-
Fuzzing: We’re currently fuzzing the EVM, consensus layer clients, etc., but there are certainly more areas to explore (e.g., the network layer).
-
Understanding risks in Ethereum’s current supply chain dependencies.
-
How to leverage LLMs (large language models) to improve protocol security (e.g., code audits, automated fuzzing tools, etc.).
Others
Question: What applications would you most like to see in the Ethereum ecosystem?
Toni Wahrstätter: To me, app developers on Ethereum have done an excellent job identifying real user needs and meeting them—even when L1 or L2 may not yet be fully ready to support certain applications. I’m particularly interested in applications combining self-custody with privacy, and there are already some great solutions. Two standout examples are Umbra and Fluidkey, both using stealth addresses to bring more privacy to everyday user interactions. Additionally, apps like Railgun, Tornado Cash, and Privacy Pools provide significant value by enhancing on-chain privacy. Back to your question, I’d love to see more wallets make privacy the default, rather than requiring users to opt in, while still delivering great UX (which is harder than people realize).
Question: Aren’t you concerned about the risk of large corporations taking over Ethereum?
Vitalik Buterin: Yes, this is absolutely an ongoing concern, and I believe the Ethereum Foundation should actively mitigate these risks. The goal is to preserve Ethereum’s neutrality, not necessarily the Ethereum Foundation’s neutrality—usually these align, but sometimes they don’t, and when they diverge, we should prioritize the former. Currently, the main risks I see are concentrated at the L2 and wallet layers, as well as staking and custody providers. The Ethereum Foundation has recently begun engaging in the first two areas, promoting adoption of interoperability standards. That said, we clearly have room to more proactively reduce risks, and we’re exploring various options.
Question: Why is the Ethereum Foundation (EF) always so opaque? So little transparency and accountability to the community.
Justin Drake: What would you like to know? The Ethereum Foundation research team holds two AMAs per year and lists all 40 researchers publicly on Research.Ethereum.Foundation. Our research is conducted openly, for example on Ethresear.ch.
Question: What are your thoughts on the future of hardware wallets?
Justin Drake: In the future, most hardware wallets will run inside phone Enclaves (rather than standalone devices like Ledger USB). With account abstraction, it’s already possible to leverage infrastructure like passkeys. I hope to see native integration (e.g., in Apple Pay) within ten years.
Vitalik Buterin: Hardware wallets need to be “truly secure” in several key aspects:
-
Secure hardware: Built on open-source, verifiable hardware stacks (e.g., see IRIS) to reduce risks of (i) intentional backdoors; (ii) side-channel attacks.
-
Interface layer security: Hardware wallets should provide sufficient transaction details to prevent the connected computer from tricking you into signing something you didn’t intend.
-
Wide accessibility: Ideally, we could make a device that serves both as a cryptocurrency hardware wallet and a secure device for other purposes, encouraging more people to actually buy and use it instead of forgetting it exists.
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
Foresight News
