Ethereum is attempting something no blockchain has ever done at scale: rewriting its core execution engine while billions of dollars continue to flow through it. In the span of a single week in late February and early March 2026, Ethereum's leadership unveiled a seven-fork "Strawmap" roadmap thro...
"Vitalik's proposals are intended to start conversations and encourage progress in difficult research areas. Community review may refine them significantly or even reject them." — Tomasz K. Stańczak, Co-Executive Director, Ethereum Foundation
Ethereum is attempting something no blockchain has ever done at scale: rewriting its core execution engine while billions of dollars continue to flow through it. In the span of a single week in late February and early March 2026, Ethereum's leadership unveiled a seven-fork "Strawmap" roadmap through 2029, Vitalik Buterin proposed replacing the Ethereum Virtual Machine (EVM) with RISC-V, and two critical Ethereum Improvement Proposals — EIP-7864 and EIP-8025 — moved from theoretical research into active implementation programs.
This is not a routine protocol upgrade. It is the most ambitious architectural overhaul in Ethereum's history, touching the state tree, the virtual machine, the consensus mechanism, and the validator economics simultaneously. The gas limit is set to increase from 60 million to potentially 200 million. Validators are being asked to shift from re-executing every transaction to verifying zero-knowledge proofs. And the entire cryptographic foundation is being hardened against quantum computing threats.
For investors, builders, and institutions, the question is not whether Ethereum needs this surgery — the network processes roughly $233 billion in market capitalization and anchors over $15 billion in tokenized real-world assets. The question is whether it can perform open-heart surgery on a system this large without killing the patient.
On February 26, 2026, Ethereum Foundation researchers published the most ambitious development roadmap the network has produced since the Merge. Dubbed the "Strawmap" — a deliberate nod to its provisional nature — the document outlines seven hard forks through 2029, organized around five "North Star" objectives: near-instant finality, dramatically higher throughput, built-in privacy, quantum-resistant security, and tighter Layer 2 integration.
The finality target alone represents a paradigm shift. Ethereum currently takes approximately 16 minutes to finalize a transaction. The Strawmap targets 8-second finality through a new consensus mechanism called "Minimmit," combined with shorter block slot times. For institutional adopters — banks settling tokenized bonds, clearinghouses processing real-world asset transfers — this is the difference between Ethereum being a settlement curiosity and a genuine financial infrastructure layer.
Throughput targets are equally aggressive. The roadmap envisions roughly 10,000 transactions per second on Layer 1, with Layer 2 networks scaling to as much as 10 million TPS. These are not aspirational numbers; they are engineering targets backed by specific EIPs and scheduled fork milestones.
Perhaps most provocatively, the Strawmap contemplates native "shielded" transfers at the base layer — allowing ETH to move without exposing full transaction details publicly. And the post-quantum cryptography push aims to replace the elliptic curve math securing the network today with schemes resilient to quantum computing attacks. The Ethereum Foundation's architecture team has committed to updating the Strawmap quarterly based on community feedback and research progress.
On March 1, 2026, Vitalik Buterin published a proposal that stunned even seasoned Ethereum developers: replace the Ethereum Virtual Machine — the bytecode execution engine that has defined Ethereum's developer experience since 2015 — with RISC-V, an open-source instruction set architecture originally designed for academic processor research.
The logic is rooted in zero-knowledge proof economics. Zero-knowledge provers are already written in RISC-V, creating natural alignment between the proving infrastructure and the execution layer. According to Buterin, the state tree and the VM together account for more than 80% of the bottleneck in efficient proving. Switching to RISC-V natively could improve execution efficiency by up to 100x and dramatically simplify proof generation.
Buterin proposed a three-stage deployment: first, using RISC-V only for precompiled contracts; second, allowing developers to deploy RISC-V contracts alongside EVM contracts; and third, retiring the EVM entirely by converting it into a smart contract running on the new RISC-V VM. Existing EVM contracts would continue to function indefinitely through this compatibility layer.
The developer community reaction has been sharply divided. Arbitrum's team disagrees with the approach, advocating instead for WebAssembly (WASM) as the contract delivery format while using RISC-V only for proofs, citing flexibility and existing hardware support. Ethereum researcher levs57 pushed back harder, arguing that the proposal rests on shaky assumptions about how proof systems work — noting that current zk-EVM implementations already rely heavily on precompiles for computationally intensive operations like hashing and signatures, meaning a VM swap might not reduce complexity as much as advertised.
Buterin himself acknowledged there is no broad consensus on the VM change. But the Ethereum Foundation's co-executive director Tomasz K. Stańczak framed the proposal as a conversation starter — standard practice for Ethereum's research-driven governance model.
If the RISC-V proposal is speculative surgery, EIP-7864 is the procedure already being prepped. This proposal targets what Buterin calls Ethereum's primary bottleneck for proof efficiency: the hexary keccak Merkle Patricia Tree used to store the network's state.
EIP-7864 replaces this structure with a unified binary state tree. The technical implications are substantial: Merkle proof sizes shrink by approximately 75%, and the number of tree branches drops by 3-4x. The current implementation uses BLAKE3 as the hash function — chosen to reduce friction for execution-layer clients experimenting with the EIP — though the final hash function remains under discussion, with Keccak and Poseidon2 as potential candidates.
The migration strategy is pragmatic. The new binary tree starts empty, with only new state changes stored in it. The existing Merkle Patricia Tree is frozen but continues to exist, allowing legacy state to be accessed without migration. This is a critical design choice: it avoids the need for a catastrophic state migration that could disrupt the network.
The EIP has moved beyond the theoretical stage. On the Ethereum Magicians forum, the proposal has attracted detailed technical review, and the Ethereum Foundation has begun scheduling breakout calls with client implementation teams.
EIP-8025 represents the most consequential change to Ethereum's validator economics since the transition to proof of stake. The proposal introduces "optional execution proofs" on the consensus layer, allowing a class of validators called "zkAttesters" to confirm blocks using cryptographic zero-knowledge proofs rather than re-executing every transaction.
The security model is conservative: a validator would only accept a block as valid after successfully verifying proofs from three out of five independent execution-layer client implementations. The shift is optional — no existing validator is forced to upgrade, and all current nodes continue to operate as they do today.
The economic implications are profound. Validators using proof-based verification would no longer need to store execution-layer state or process every transaction. This dramatically lowers hardware requirements — potentially from enterprise-grade servers to consumer-grade machines — and reduces the time needed to sync with the network. The Ethereum Foundation projects that 1 in 10 validators could be verifying ZK proofs by the end of 2026.
The infrastructure cost question is non-trivial. Generating ZK proofs requires specialized hardware — high-end GPUs or ASICs that can cost $100,000 or more for proof clusters. However, the ZKsync Airbender team has demonstrated the ability to prove Ethereum blocks in under 50 seconds using a single GPU, suggesting that proving costs are falling faster than expected. The publication of a concrete 2026 roadmap, scheduled breakout calls with project leads, and the drafting of an EIP with specific peer-to-peer gossip mechanisms signal that this work has transitioned from research into an active implementation program.
The Strawmap's ambitions crystallize into two concrete upgrades scheduled for 2026. Glamsterdam, expected in the first half of the year, bundles up to 22 EIPs focused on Layer 1 scalability. The headline feature is parallel transaction processing — overhauling execution so nodes can handle multiple independent chunks of state simultaneously, breaking the linear pipeline that currently bottlenecks throughput.
A key technical innovation is "Block-Level Access Lists," which pre-declare the accounts and smart contracts a block will interact with. This reduces computational overhead for nodes and lays the groundwork for parallel verification. The upgrade also introduces a critical decoupling of state creation from execution gas, allowing compute to scale massively without bloating the state.
The gas limit target tells the story in a single number. The Ethereum Foundation aims to push the limit from the current 60 million "toward and beyond" 100 million in the first half of 2026. Following enshrined proposer-builder separation (ePBS) — which Hegota is expected to deliver — the limit could double to 200 million. One analysis projects this could cut gas fees by up to 78%.
Hegota, slated for the second half of 2026, targets higher gas limits, native account abstraction, and post-quantum readiness. Together, the two forks represent the most feature-dense upgrade year Ethereum has attempted since the Merge.
Ethereum trades at approximately $1,975 as of March 7, 2026 — down more than 60% from its August 2025 all-time high of $4,953. The correction has been driven overwhelmingly by macro forces rather than any deterioration in network fundamentals. This creates a peculiar situation: Ethereum is executing the most ambitious technical roadmap in its history while its market valuation reflects deep pessimism.
The economic value redistribution from these upgrades is significant. If ZK-based block verification succeeds, the cost structure for running a validator changes fundamentally. Lower hardware requirements could democratize validation, increasing decentralization — but they could also compress validator yields if the barrier to entry drops enough to flood the network with new validators. The proving infrastructure creates a new economic layer: someone must pay for proof generation, and the market for ZK proving services is emerging as a distinct infrastructure category.
For Layer 2 networks, Ethereum's scaling ambitions are a double-edged sword. Higher L1 throughput and lower fees reduce the urgency of L2 usage for cost savings — potentially compressing L2 revenue models. But tighter L2 integration, as outlined in the Strawmap, could make rollups more valuable as specialized execution environments rather than mere cost arbitrage layers.
The privacy features contemplated in the Strawmap could unlock entirely new use cases — institutional treasury management, compliant but private settlement, and shielded payroll — that Ethereum currently cannot serve. Real-world assets tokenized on Ethereum have already crossed $15 billion; native privacy could accelerate this dramatically.
The Strawmap is Ethereum's most ambitious roadmap since the Merge, outlining seven hard forks through 2029 targeting 8-second finality, 10,000 L1 TPS, native privacy, and quantum resistance.
The RISC-V proposal is controversial but directionally important. Even if the community rejects a full VM swap, the conversation has forced Ethereum to confront the EVM's limitations as a proving bottleneck — and alternatives like WASM are now actively debated.
EIP-7864 and EIP-8025 have moved from research to implementation. The binary state tree migration and optional ZK proof verification are no longer theoretical; they have scheduled breakout calls, client implementation teams, and concrete EIP drafts.
Validator economics are about to change. The shift from transaction re-execution to ZK proof verification lowers hardware requirements but introduces new cost structures around proof generation infrastructure.
The gas limit trajectory from 60M to 200M could reshape fee economics across the entire Ethereum ecosystem, compressing Layer 2 cost advantages while unlocking new L1 use cases.
Ethereum is trading at a 60% discount to its all-time high while executing the most technically ambitious year in its history — a divergence that will resolve in one direction or another.
What Ethereum is attempting in 2026 has no precedent in blockchain engineering. The network is proposing to replace its virtual machine, restructure its state storage, change how validators verify blocks, double its gas capacity, and lay the groundwork for quantum resistance and native privacy — all while securing hundreds of billions in assets and anchoring the largest decentralized application ecosystem in existence.
The risks are proportional to the ambition. A botched upgrade could fracture the developer community. A RISC-V migration that breaks backward compatibility could strand billions in existing smart contracts. A premature shift to ZK-based validation could introduce novel attack vectors.
But the alternative — standing still — is arguably riskier. Solana is shipping aggressive infrastructure upgrades. Alternative L1s are competing on throughput and cost. And institutional adoption requires the kind of finality, privacy, and throughput guarantees that Ethereum cannot deliver today.
The Strawmap is Ethereum's bet that the network effect and developer ecosystem built over a decade are durable enough to survive radical surgery. The next 12 months will determine whether that bet pays off.