On February 18, 2026, the Ethereum Foundation published its most consequential protocol roadmap since the Merge. Organized around three newly defined development tracks — Scale, Improve UX, and Harden the L1 — the update targets two back-to-back hard forks in 2026: **Glamsterdam** in the first ha...
"Write privacy is actively destroying value today. It's hundreds of millions per year being extracted from users because their transaction intent is visible before execution." — Tanisha Katara, Katara Consulting
On February 18, 2026, the Ethereum Foundation published its most consequential protocol roadmap since the Merge. Organized around three newly defined development tracks — Scale, Improve UX, and Harden the L1 — the update targets two back-to-back hard forks in 2026: Glamsterdam in the first half and Hegotá in the second. The ambition is staggering: push the gas limit from 60 million toward and beyond 100 million, enshrine proposer-builder separation directly into the protocol, lay groundwork for parallel execution, and begin the long march toward post-quantum cryptographic security.
At the center of Glamsterdam sits EIP-7732 (enshrined Proposer-Builder Separation), a structural reform that would eliminate the trust assumptions embedded in today's off-chain MEV relay infrastructure — a system through which nearly $24 million was extracted from users in a single 30-day period between December 2025 and January 2026. Combined with Block-Level Access Lists (EIP-7928) and gas repricing (EIP-7904), Glamsterdam represents not a routine upgrade but an economic redesign of how value flows through Ethereum's base layer. For an ecosystem generating approximately $1 billion annually in MEV extraction across major chains, the stakes could not be higher.
The 2026 roadmap marks a structural departure from how Ethereum protocol development has been organized. Previously, work was siloed into tightly scoped milestones. Now, the Ethereum Foundation has consolidated its efforts into three parallel tracks, each with dedicated leadership:
Scale — Led by Ansgar Dietrichs, Marius van der Wijden, and Raúl Kripalani. This track merges what was previously split between L1 execution scaling and blob scaling into a unified effort. The explicit target: push gas limits "toward and beyond 100 million" while continuing to expand blob capacity for Layer 2 data availability. The track also owns the path toward zero-knowledge proof attester clients — a prerequisite for validators to shift from re-executing every block to verifying ZK proofs.
Improve UX — Led by Barnabé Monnot and Matt Garnett. The focus is native account abstraction (EIP-7701 and EIP-8141, known as Frame Transactions) and cross-L2 interoperability through the Open Intents Framework. The goal: make smart contract wallets first-class citizens without relayers, and streamline the fragmented experience of moving assets across rollups.
Harden the L1 — Led by Fredrik Svantes, Parithosh Jayanthi, and Thomas Thiery. This is an entirely new track, reflecting growing urgency around post-quantum cryptographic preparedness, censorship resistance (FOCIL, EIP-7805), and testing infrastructure hardening. The creation of a dedicated security track signals that Ethereum's leadership views resilience not as a feature but as a precondition for everything else.
This reorganization came alongside a leadership transition: Tomasz Stańczak stepped down as co-executive director at month's end, with Bastian Aue assuming interim duties.
Glamsterdam — a portmanteau of "Gloas" (consensus layer) and "Amsterdam" (execution layer) — is targeted for May or June 2026. It comprises up to 22 Ethereum Improvement Proposals focused on L1 scalability and MEV reform. The scope freeze is set for late Q1 2026, after which no new EIPs can be added.
The three headliner proposals define the upgrade's character:
| EIP | Name | Purpose | |-----|------|---------| | EIP-7732 | Enshrined PBS (ePBS) | Embeds proposer-builder separation into the protocol, removing reliance on off-chain relays | | EIP-7928 | Block-Level Access Lists | Enables blocks to pre-declare data access patterns, laying plumbing for parallel execution | | EIP-7904 | Benchmarked Gas Repricing | Realigns gas costs to match actual computational resource consumption |
The significance of shipping these three together cannot be overstated. Individually, each addresses a known bottleneck. Together, they reconfigure the economic plumbing of block production, execution efficiency, and resource pricing in a single coordinated move.
Today, Ethereum's block production relies on an off-chain relay system operated primarily through Flashbots' MEV-Boost infrastructure. Validators (proposers) outsource block construction to specialized builders who compete to assemble the most profitable blocks. The relay sits between them, trusted to faithfully convey bids without tampering. This works — but it introduces centralization risk, trust assumptions, and censorship vectors.
The numbers are sobering. Between December 8, 2025, and January 6, 2026, nearly $24 million in MEV profit was extracted on Ethereum — approximately $800,000 per day. In 2025, sandwich attacks alone constituted $289.76 million of the $561.92 million in total MEV transaction volume, representing 51.56% of all extraction. On Layer 2 networks, the situation is arguably worse: MEV search spam consumes over 50% of gas on major L2s including Unichain and OP Mainnet while paying under 10% of fees.
EIP-7732 fundamentally restructures this dynamic. Under ePBS:
Research indicates ePBS could reduce MEV extraction by up to 70%. However, the transition introduces its own risks. An academic paper on the "free option problem" estimates that under an 8-second option window, builders could exercise options on approximately 0.82% of blocks under normal conditions, rising to ~6% during high-volatility periods. This means the design must balance MEV reduction against liveness concerns during market stress.
As Andy Guzman of the Ethereum Foundation's Privacy and Scaling team noted, private writes — shielding transaction intent before execution — is "the first thing you have to do" to address the MEV problem structurally. ePBS is the protocol-level enforcement of that principle.
Ethereum's gas limit has been one of the most conservative parameters in crypto. It sat at 30 million from 2021 until early 2025, when validators coordinated to double it to 60 million. Now the Foundation wants to push it "toward and beyond" 100 million — a move that would fundamentally alter Ethereum's throughput characteristics.
The scaling math is straightforward:
| Scenario | Gas Limit | Gas/Second | Simple Tx/Sec (21k gas) | Complex Tx/Sec (120k gas) | |----------|-----------|------------|------------------------|--------------------------| | Current | 60M | 5.0M | ~238 | ~42 | | 2× current | 120M | 10.0M | ~476 | ~83 | | High-end target | 200M | 16.7M | ~793 | ~139 |
For DeFi protocols, higher gas limits translate directly to tighter DEX spreads, faster liquidations, and the execution density required for sophisticated strategies like high-frequency on-chain market making. The block-level access lists introduced by EIP-7928 are critical infrastructure for this scaling: by allowing blocks to pre-declare which accounts and storage slots they will access, clients can preload data and execute transactions in parallel rather than sequentially.
But gas limit increases are not free. The structural bet behind very high gas limits is validator ZK-proof adoption — shifting from block re-execution to proof verification. If proof generation becomes dominated by a few specialized entities with expensive hardware, Ethereum's security model, which depends on distributed validation, could weaken. This is the hidden validator risk embedded in the 2026 roadmap.
Layer 2 networks are not passive observers of L1 upgrades. Base, Coinbase's L2, published a detailed analysis of the ~50 non-headliner EIPs proposed for Glamsterdam, identifying six as most strategically valuable. These fall into three categories:
Blob Scaling — Expanded blob capacity directly reduces L2 data availability costs, enabling continued throughput growth. Fusaka's PeerDAS implementation in December 2025 already delivered an 8× theoretical increase in blob capacity; Glamsterdam aims to push further.
Gas Repricing — EIP-2780 proposes reducing the intrinsic base cost of a transaction from 21,000 to 4,500 gas, making simple ETH transfers dramatically cheaper for L2 users while fitting more transactions per block.
Builder UX — Improvements to how block builders interact with L2 settlement transactions, reducing friction in the posting of rollup batches to L1.
Critically, L2 networks like Linea are not waiting for L1 upgrades. Linea increased its own block size limit to 60 mGas in December 2025 and is targeting further increases throughout 2026. The question for Ethereum is whether L1 scaling can complement — rather than lag behind — the rollups it is supposed to anchor.
Glamsterdam is only Act One. Hegotá, planned for H2 2026, extends the roadmap into three dimensions:
The proposal window for Hegotá ran from January 8 to February 4, 2026, and the scope is expected to solidify by mid-year. If Glamsterdam is the economic restructuring, Hegotá is the security hardening — acknowledging that quantum computing threats require proactive protocol-level response, not reactive patches.
Together, the two forks represent the most ambitious single-year protocol evolution Ethereum has attempted since the Merge in 2022.
From an economic value distribution perspective, Glamsterdam reshapes who captures what across Ethereum's stack:
Validators face a complex transition. ePBS standardizes MEV flows, potentially compressing the premium that sophisticated validators earn from optimized block building. Simultaneously, gas limit increases generate more total fees per block — but only if demand scales to fill the new capacity. The net effect on validator economics is ambiguous and will depend heavily on post-fork demand dynamics.
MEV searchers and builders face an existential restructuring. Enshrined PBS removes the relay layer that current builders depend on, creating a native auction mechanism. Builders who can assemble the most gas-efficient, value-dense blocks will thrive; those reliant on information asymmetries created by the relay system may see margins compress.
Users stand to benefit most directly. Reduced MEV extraction, lower gas costs from repricing, and higher throughput should translate into better execution prices across DeFi. The private transfer gas premium — currently 420,000 gas versus 21,000 for public transfers, a 20× penalty — highlights how far the protocol still needs to go to make privacy economically viable at scale.
L2 operators gain cheaper data availability and more efficient settlement. For L2 ecosystems collectively processing billions in daily volume, even marginal improvements in L1 posting costs compound significantly.
Glamsterdam is Ethereum's most significant economic upgrade since the Merge, targeting MEV reform, execution scaling, and gas repricing in a single coordinated fork slated for May/June 2026.
ePBS (EIP-7732) eliminates the off-chain relay system that currently mediates $800,000/day in MEV extraction, enshrining proposer-builder separation directly into the protocol.
The gas limit push from 60M toward 100M+ depends on a structural shift from block re-execution to ZK-proof verification — a transition that carries meaningful validator centralization risk.
L2 networks are actively shaping Glamsterdam's scope, with Base identifying six priority EIPs including a proposal to cut base transaction costs by nearly 80%.
Hegotá in H2 2026 extends the vision with native account abstraction and post-quantum cryptographic readiness, making 2026 a potential inflection year for Ethereum's protocol maturity.
The economic value redistribution is real: MEV flows get restructured, validator revenue models shift, and users stand to recapture hundreds of millions in currently extracted value.
Ethereum's 2026 roadmap is not an incremental upgrade cycle. It is a coordinated restructuring of the protocol's economic architecture — who builds blocks, how value is extracted, what transactions cost, and how the network scales. Glamsterdam tackles the immediate economic distortions created by the MEV relay system and gas mispricing. Hegotá addresses the longer-term existential questions of quantum resistance and account abstraction.
The risk is execution. Shipping ePBS, block-level access lists, gas repricing, and up to 22 EIPs in a single fork is technically ambitious. The validator centralization risk inherent in ZK-proof adoption is real and underappreciated. And the competitive pressure from alternative L1s and from Ethereum's own L2 ecosystem creates a narrow window for the base layer to prove its continued economic relevance.
But if the Ethereum Foundation delivers on even 80% of this roadmap, 2026 will be the year Ethereum transitions from a scaling narrative to a scaling reality — with economic implications that ripple across every protocol, validator, and user in the ecosystem.