“We need a new path.” With that blunt assessment, Ethereum co-founder Vitalik Buterin has signaled a sharp turn in how he believes the network should scale—away from heavy reliance on layer-2 (L2) solutions and toward a more balanced, protocol-centered approach.
In a recent post on X, Buterin argued that the “original vision” for L2 networks and their place in Ethereum’s roadmap “no longer makes sense.” For years, the dominant narrative in the Ethereum ecosystem has been that rollups and other L2s would shoulder most of the load for scaling, with the main chain (layer 1, or L1) acting as a secure settlement and data-availability layer. Now, Buterin is openly questioning whether that strategy, as currently implemented, actually serves Ethereum’s core values.
At the heart of his critique is decentralization. Buterin pointed out that several L2 networks have made notable compromises on decentralization—often in the form of centralized sequencers, upgrade keys controlled by small teams, or governance structures that concentrate power. These systems are still technically “on Ethereum,” but he argued they should not automatically be branded or perceived as direct extensions of Ethereum’s own trust guarantees.
This distinction matters. In theory, L2s were supposed to inherit Ethereum’s security while adding throughput and lower fees. In practice, many of them introduce additional trust assumptions: users may have to rely on a centralized operator to process transactions, a multisig to manage upgrades, or a small committee to resolve disputes. Buterin’s message is that the ecosystem should be more honest and precise about what these systems actually are—and what they are not.
The long-standing vision for Ethereum scaling has been clear: increase effective block space so that more applications, users, and transactions can coexist without sacrificing security, efficiency, or decentralization. Rollups were a big part of that story. They promised to move computation off-chain while anchoring critical data and proofs on Ethereum, supposedly delivering the best of both worlds. But as L2s have grown in number and economic importance, the trade-offs they made to launch quickly and capture market share have become harder to ignore.
Buterin highlighted two key tensions in current discussions around L2s. First, the slow and uneven progress toward truly decentralized “stage 2” rollups—systems that are not just technically integrated with Ethereum but also robustly permissionless and trust-minimized. Second, ongoing interoperability issues and fragmentation between different L2 environments. Instead of a cohesive scaling layer, the ecosystem is drifting toward a patchwork of partially independent platforms with varying security and governance properties.
For users, this creates a confusing landscape. A transaction on one L2 may not carry the same guarantees as a transaction on another, even if both market themselves as “Ethereum L2s.” Bridges between L2s introduce additional risk. Liquidity and applications get split across multiple chains. The simple mental model—“it’s all just Ethereum”—no longer holds up cleanly.
Buterin’s call for a “new path” is therefore not a rejection of L2s entirely, but a warning against over-optimizing the roadmap around them. He is effectively arguing that Ethereum should avoid becoming overly dependent on external scaling layers whose incentives, governance, and technical decisions may diverge from Ethereum’s core principles. Instead, the base layer should play a more active role in defining standards, setting expectations, and delivering meaningful scaling directly at L1.
One part of this shift is a renewed focus on making Ethereum’s core protocol more powerful, rather than assuming that most innovation and scaling will happen on L2. That could mean prioritizing features that directly improve user experience and security at the base layer—such as better account abstractions, more efficient data-availability mechanisms, and stronger enshrined cryptographic primitives that simplify the design of rollups and other extensions.
Another part is clearer categorization. Buterin’s criticism that some systems should not be “branded” as Ethereum extensions is a call for sharper language and marketing discipline. Not every chain that posts data to Ethereum should be implicitly treated as if it offers Ethereum-level decentralization and censorship resistance. The ecosystem may need explicit labels that distinguish between fully trust-minimized rollups, partially trusted L2s, appchains with looser security models, and sidechains that simply use Ethereum as a funding or bridging hub.
In practical terms, this also raises the bar for what counts as an “Ethereum-aligned” L2. The aspirational target remains what many call a “stage 2” rollup: no centralized escape hatches, robust fault or validity proofs, permissionless participation, and minimal governance over core security properties. Buterin’s remarks suggest that until a system reaches something close to that standard, it should be treated more like an independent platform building on Ethereum, not as a direct continuation of Ethereum itself.
For developers and entrepreneurs, this pivot forces a reassessment of strategy. Building solely around a hot L2 ecosystem may no longer be enough to claim deep alignment with Ethereum’s roadmap. Teams will increasingly be judged on how they handle decentralization, upgradeability, and user sovereignty, not just on fees and throughput. Those that rely heavily on centralized sequencers or opaque governance may find themselves under sharper scrutiny from both users and core researchers.
Users, too, will need to become more discerning. Rather than asking only which chain has the lowest fees or fastest confirmations, they will have to consider who controls the infrastructure: Can withdrawals be censored? Who can pause the network or push emergency upgrades? How easy is it for new validators or provers to join? Buterin’s shift in rhetoric signals that such questions should be front and center when evaluating any L2 that claims to be “part of Ethereum.”
This moment also reopens a broader philosophical debate: Should Ethereum be primarily a minimal, neutral base layer with most complexity pushed outward to L2s and application layers, or should it take a more opinionated stance, enshrining certain scaling primitives and standards directly into the protocol? Buterin’s call for a new path leans toward the latter—toward a base layer that more actively shapes the environment in which L2s operate, rather than simply serving as passive infrastructure.
We may see more experimentation with “enshrined” mechanisms: rollup-friendly protocol features, native data-availability tools, or standardized interfaces that reduce the temptation for L2s to centralize critical components. The goal would be to make the most decentralized option also the most practical, thereby aligning economic incentives with Ethereum’s long-term values.
At the same time, the ecosystem is unlikely to abandon L2s altogether. They remain an essential tool for scaling, especially for high-frequency applications like trading, gaming, and payments. The emerging roadmap is less about replacing L2s and more about redefining their role: from being the centerpiece of Ethereum’s scaling story to being one of several coordinated layers in a broader, more principled architecture.
In that redesigned architecture, Ethereum’s L1 would be more than a mere settlement layer; it would be the anchor of trust that actively shapes what it means to be an “Ethereum-based” chain. L2s that align closely with that anchor—by minimizing trust assumptions and decentralization compromises—will likely be treated as first-class citizens. Others may still thrive as independent ecosystems, but without the implicit halo of Ethereum-level security.
Buterin’s intervention is therefore both a technical and political move. Technically, it challenges developers to rethink the stack and push for deeper protocol improvements. Politically, it asks the community to be more honest about trade-offs, more precise with branding, and more protective of Ethereum’s identity as a credibly neutral, decentralized platform.
How quickly this new path materializes will depend on concrete roadmap updates, coordination among client teams, and the willingness of major L2 projects to evolve. Some may double down on decentralization to meet higher expectations; others may embrace their status as semi-independent ecosystems building around Ethereum rather than inside it.
What is clear from Buterin’s latest comments is that the era of uncritical L2 evangelism is over. Scaling, in his view, cannot be achieved by outsourcing core values to external networks. If Ethereum is to remain Ethereum—secure, efficient, and genuinely decentralized—it must re-center its roadmap on the base layer, redefine how L2s fit into that vision, and chart a more nuanced path forward than simply “move everything off-chain.”