EIP-7928 parallelization, native privacy roadmap, EIP-8141 deep dive, EF restructuring

Welcome to our weekly digest, where we unpack the latest in account and chain abstraction and the broader infrastructure shaping Ethereum.

This week: EIP-7928 and how Block-level Access Lists bring parallel execution to Glamsterdam; Vitalik maps Ethereum’s near-term native privacy roadmap around AA, FOCIL, and keyed nonces; Alex Forshtat delivers a deep dive on EIP-8141 frame transactions at ETHPrague; and Vitalik publishes his perspective on a leaner, more focused Ethereum Foundation.

EIP-7928: Block-level Access Lists Bring Parallel Execution to Glamsterdam
Mapping Ethereum’s Near-Term Native Privacy Roadmap
EIP-8141 Frame Transactions: Deep Dive from ETHPrague
Vitalik Outlines a Leaner, More Focused Ethereum Foundation

Please fasten your belts!

EIP-7928: Block-level Access Lists Bring Parallel Execution to Glamsterdam

EF researcher Toni Wahrstätter published a detailed thread breaking down EIP-7928, the execution-layer headliner for Glamsterdam alongside ePBS. The proposal adds a Block Access List (BAL) to every Ethereum block — a list of every account and storage slot the block touches, plus post-transaction state diffs showing exactly what changed after each transaction.

With state diffs available upfront, client software no longer has to discover storage locations while executing. Everything needed can be prefetched into cache in parallel before execution begins, resolving transaction dependencies in advance and enabling true parallel execution across Geth, Nethermind, Besu, Erigon, Reth, and others.

The gains are already measurable. On 6-core machines, EL clients validate blocks up to 5x faster with BALs in place. Combined with ePBS decoupling the block from the payload, giving validators 2–4x more execution time, Glamsterdam will ship with a 200M gas limit, with 300M described as a realistic near-term follow-up. Snap sync also improves: BALs let nodes skip the healing phase entirely when catching up to the chain tip.

For AA builders, higher throughput and cheaper state access directly reduce the cost of batched UserOps, gasless flows, and cross-chain intents at scale. EIP-7928 is one of the clearest signs yet that Ethereum’s base-layer scaling is moving from theory into production.

Mapping Ethereum’s Near-Term Native Privacy Roadmap

Vitalik Buterin outlined three near-term steps to make privacy a native property of Ethereum. The post extends the cypherpunk reset he outlined in January and the privacy framing he introduced at the Web3 Festival in Hong Kong earlier this spring.

The first initiative combines account abstraction with FOCIL. Right now, private transactions pass through the public mempool where block builders can see and exclude them. FOCIL changes this by allowing a committee of validators to propose transaction inclusion lists that builders are required to honour — or risk having their block rejected. Combined with AA, private transactions gain first-class inclusion guarantees at the protocol level.

The second step is keyed nonces, proposed in EIP-8250. Ethereum’s current sequential nonce model creates a bottleneck when multiple private transactions need to execute in parallel from the same account. Keyed nonces let each spend use its own nonce domain, including one derived from a privacy nullifier, making transactions on different keys replay-independent.

The third is access-layer work, centred on the Kohaku privacy toolkit and private reads. This targets the metadata layer: hiding users’ wallet queries from centralised node providers so that RPC calls don’t leak which addresses a user is watching. Buterin framed all three as essential for giving ETH true “moneyness” and argued that L1 privacy could drive more activity back to Ethereum mainnet. None of the three are live yet.

EIP-8141 Frame Transactions: Deep Dive from ETHPrague

At ETHPrague 2026, ERC-4337 core developer Alex Forshtat delivered a technical deep dive on EIP-8141, shared this week by the @erc4337 account. The talk is the clearest walkthrough yet of why frame transactions represent an endgame account abstraction design rather than another incremental layer.

The core change: instead of a single top-level execution frame, a frame transaction contains multiple frames each with an explicit role modifier. The VERIFY frame, which must end with an APPROVE opcode, handles validation and signals to the protocol that if this frame reverts, the transaction cannot be included in a valid block at all. This separates validation from execution at the protocol level, replacing the fixed ECDSA + nonce + balance check with fully programmable logic.

For builders, the practical implications are significant. Frame transactions natively support gas abstraction via a canonical paymaster, atomic batching across multiple execution frames, passkey signatures, and full backward compatibility with existing EOAs — which receive a canonical default validation code automatically. EIP-8141 also provides a clean path to post-quantum signatures: the VERIFY frame decouples transaction authorization from ECDSA entirely, so accounts can adopt quantum-safe schemes without changing their address.

EIP-8141 currently holds Considered for Inclusion status for the Hegotá fork (H2 2026). Forshtat addressed the mempool DoS risk directly: a mempool validation strategy restricts storage reads during validation to the sender’s own storage and caps verification gas at ~100,000 units, with the canonical paymaster required for public mempool propagation. The full talk is worth watching for any team thinking seriously about the native AA transition.

Vitalik Outlines a Leaner, More Focused Ethereum Foundation

On May 24, Vitalik Buterin published a lengthy post on X addressing months of turbulence at the Ethereum Foundation, defending its restructuring and laying out his personal view of the EF’s long-term direction. The post came after eight senior EF contributors left or announced departures in 2026, five of them in May, including all three Protocol Cluster leads.

Buterin framed the EF’s mandate around what he called the CROPS dimension: censorship resistance, capture resistance, openness, privacy, and security. He argued that maximising throughput and speed at the expense of these properties leads to mediocrity, and that Ethereum should be “deeply impressive” on CROPS rather than competing on execution speed alone. He also noted that the EF holds only 0.16% of all ETH, far below the 10–50% typical for other blockchain foundations, and committed to reducing ETH sales to prioritise longevity.

On governance, Buterin confirmed that interim co-executive director Bastian Aue is leading the transition, that the board is expanding, and that his own influence within the EF will continue to decrease — which he described as the outcome he wants. He also pointed to AI-assisted formal verification as a near-term path to making Ethereum provably bug-free.

For the AA ecosystem, the CROPS framing matters directly. Privacy and censorship resistance, two of the five pillars, map precisely onto the work covered elsewhere in this issue: the native privacy roadmap, FOCIL, and EIP-8141’s design goals around uncensorable transaction inclusion and signature agility. The EF’s stated focus, whatever the organisational turbulence, remains aligned with the infrastructure AA builders depend on.

Start exploring Account Abstraction with Etherspot!