Ethereum co-founder Vitalik Buterin has laid out what he calls a “plausible roadmap” for addressing one of the network’s most persistent challenges: high data costs and limited scalability. In a blog post titled Endgame, Buterin discussed how Ethereum could improve throughput while preserving the properties many users and developers consider essential, especially trustless validation, decentralization, and censorship resistance.
The post arrives against the backdrop of recurring criticism aimed at Ethereum’s cost structure and scaling limits. Rather than presenting a simple short-term fix, Buterin described a broader architectural direction for how the network might evolve over time, including the role of the proof-of-stake transition, the use of rollups, and mechanisms designed to prevent block producers from exerting too much control over which transactions make it on-chain.
A roadmap built around scalable validation
In Endgame, Buterin outlined several concepts that could form part of Ethereum’s future design. Among them is a “second tier” of staking with low resource requirements, along with the introduction of either fraud proofs or ZK-SNARKs so that ETH users can obtain block validity more cheaply. The broad objective is not only to scale the network, but to do so without sacrificing the core security guarantees that define Ethereum’s long-term vision.
Buterin described a possible end state in which block production remains centralized to some degree, while block validation is trustless and highly decentralized. On top of that, he argued for specialized anti-censorship mechanisms intended to limit the ability of block producers to filter or suppress transactions. In his view, the result may not be aesthetically elegant, but it could still preserve the guarantees that matter most.
He was explicit about the tradeoffs. Even in a scenario where all major stakers or block producers attempted to attack the network or engage in censorship, Buterin argued that the worst-case outcome would be that they all go offline, causing the chain to stop accepting transactions until the broader community coordinates resources and brings at least one honest primary-staker node online. That framing underscores the central idea of the post: Ethereum may end up with messy or hybrid structures, but those structures can still be engineered to remain robust under adversarial conditions.
Rollups remain central to Ethereum’s scaling thesis
A major theme of Buterin’s roadmap is Ethereum’s continued commitment to a rollup-centric future. His comments follow prior discussions among Ethereum developers around EIP-4488, a proposal associated with lowering data costs for rollups. According to the source material, the plan could reduce data transfer costs to roughly one-fifth of prior levels, and Ethereum developer Tim Beiko had also shared views on how such changes might lower rollup expenses.
Buterin argued that Ethereum is especially well-positioned for an uncertain technological future precisely because the rollup-centric approach does not force the ecosystem to prematurely commit to a single winning model. Instead, it creates room for multiple scaling paths to develop in parallel. That flexibility, in his view, is one of the deepest advantages of Ethereum’s strategy.
He stressed that Ethereum is “open to all of the futures,” suggesting that the protocol’s job is not to dictate one final architecture too early, but to create a base layer capable of supporting several plausible outcomes. This point matters because the scaling debate is no longer only about raw transaction throughput. It is also about which forms of decentralization are sustainable, which types of infrastructure are realistic, and which tradeoffs users and builders are willing to accept.
Questions around decentralization and cross-domain MEV
Buterin also introduced a note of caution into the decentralization debate. He suggested that Ethereum researchers should think carefully about what level of decentralization in block production is actually achievable in practice. In his telling, it may not be worth adding highly complicated plumbing solely to make block production maximally decentralized if structural forces such as cross-domain MEV — or even cross-shard MEV from a single rollup occupying multiple shards — ultimately make that goal unsustainable.
This is a notable part of the argument because it shifts the conversation away from ideological purity and toward practical resilience. Rather than assuming every layer of the system can or should become fully decentralized in the same way, Buterin appears to be asking whether the protocol can remain credibly neutral and censorship-resistant even if some operational functions remain relatively concentrated. In other words, the design challenge is less about eliminating all centralization and more about preventing concentrated actors from undermining the network’s integrity.
That distinction is especially relevant in a multi-rollup future, where value and activity may be spread across numerous execution environments while still depending on Ethereum’s base layer for security and data availability. As those domains interact, MEV could become an increasingly important force shaping the economics and centralization pressures of the ecosystem.
What this means for “big block chains”
Beyond Ethereum itself, Buterin addressed the broader debate over so-called “big block chains”. He argued that there is still a path for such systems to evolve into models that are trustless and censorship-resistant. However, he added an important condition: whether that path is taken will depend on whether their core developers and communities truly value censorship resistance and decentralization enough to pursue it.
This comment broadens the scope of the discussion from Ethereum-specific engineering choices to a more general question facing the blockchain industry. For Buterin, scalability alone does not define success. The more important issue is whether networks can grow while preserving the properties that distinguish public blockchains from conventional infrastructure. If they cannot, then higher throughput may come at too high a cost.
A long timeline for sharding, DAS, and production-grade zk systems
Despite the forward-looking tone of the post, Buterin made clear that this roadmap is not a near-term finish line. He concluded that it will likely take years for the full picture to play out. Technologies such as sharding and data availability sampling remain highly complex to implement, and he emphasized that significant refinement and auditing will be needed before users are fully comfortable storing assets in a ZK-rollup running a full EVM.
He also noted that research into cross-domain MEV is still in its infancy. That means several of the most important assumptions behind Ethereum’s future design are still being tested, modeled, and debated. The destination may be increasingly visible, but the engineering path remains difficult.
In practical terms, Buterin’s Endgame post does not promise that Ethereum’s scaling challenges will disappear quickly. Instead, it presents a framework for understanding how the network might mature: centralized tendencies in some areas, strong decentralization in others, trustless validation, rollup-driven scale, and explicit anti-censorship protections. The resulting system may be less clean than idealized blockchain narratives suggest, but Buterin argues that it can still deliver the guarantees users care about most.
For the Ethereum community, the message is twofold. First, scalable blockchains are possible, and Ethereum has a credible route toward that future. Second, getting there will require patience, technical discipline, and a willingness to confront tradeoffs directly rather than pretend they do not exist. If Buterin’s roadmap proves directionally correct, Ethereum’s endgame will not be defined by perfect decentralization at every layer, but by the ability to maintain credible neutrality, user verifiability, and censorship resistance at scale.