Ethereum is undergoing a foundational redesign. A draft proposal, EIP-8025 (Optional Execution Proofs), rewrites the core validation logic: validators will verify compact zero-knowledge proofs instead of re-running entire blocks. This marks Ethereum's shift from transaction re-execution to proof verification.
zkEVM Team's 2026 Roadmap: Standards and Infrastructure
The Ethereum Foundation's zkEVM team outlined six priorities for 2026: setting execution witness and program standards, ensuring API compatibility, layer integration, proof infrastructure, measurement, and security. The overarching goal: make verification costs accessible and practical for everyone.
12 GPUs as the Hurdle, Centralization Risk Emerges
The new design relies on an Execution Witness, a standard program monitors state changes and produces zero-knowledge proofs, which validators use to confirm blocks. A key complementary upgrade, Enshrined Proposer-Builder Separation (ePBS), extends the time window for block verification. However, proof generation demands high-end hardware. Latest analysis shows 12 GPUs on average are needed to verify an Ethereum block. The high barrier could force small validators out, concentrating power among large providers. The project's security assumption hinges on three independent proofs from diverse software clients — a test for decentralization.
Statelessness: Bringing Home Validators Back In
Ethereum's updated roadmap emphasizes statelessness — blocks can be verified without carrying the full state. This promises drastically reduced sync times, allowing home validators to fully experience network participation. Even with rising gas limits, stable verification costs could accommodate more transactions.
L2s Must Pivot: Scale Alone No Longer a Moats
If Layer 1 verification costs drop and infrastructure is shared, Layer 2 projects that only boast scalability will struggle to stand out. Future L2 differentiation is expected to come from specialized VMs, ultra-low latency, novel interaction models, among other features.
Three Medium-to-Long-Term Scenarios
The draft sketches three possible paths. First, widespread proof verification lowers entry barriers for home users. Second, high hardware requirements for proof production accelerate centralization, favoring large providers. Third, shared L1 proof infrastructure pushes L2 projects to seek new niches. Success depends on harmonizing technical standards, lowering hardware requirements, and stabilizing metrics. Upgrades like ePBS and Glamsterdam may further widen the verification window, enhancing overall network effectiveness. The proposal remains in early draft phase, not expected to ship in the near term. Priorities include testing, maturing the approach, and transparently understanding the new architecture.