
The Ethereum Foundation identified EIP-7805 (FOCIL) and EIP-8141 (Frame Transactions) as two 'must ship' proposals for the Hegotá upgrade, prioritizing censorship resistance and flexible account authentication.
AI-generated summary
Ethereum's last upgrade, Fusaka, launched Dec. 3, 2025, while the upcoming Glamsterdam upgrade focuses on base-layer scalability.
The Ethereum Foundation has identified two “must ship” proposals for the Hegotá upgrade, prioritizing censorship resistance and more flexible account authentication as it narrows the fork’s scope.
The ranking evaluated 62 proposed Ethereum Improvement Proposals (EIPs), drawing on input from roughly 60 researchers and engineers across the Foundation’s Protocol cluster. It marks the first time the Foundation’s Protocol cluster has published a unified view of proposed EIPs rather than separate opinions from individual teams.
The two must-ship proposals are EIP-7805, known as FOCIL, and EIP-8141, or Frame Transactions, which the Foundation designated as the consensus and execution-layer headliners for Hegotá.
FOCIL is designed to give users a path to have eligible transactions included without relying on centralized block builders, strengthening Ethereum’s resistance to transaction censorship.
Frame Transactions would introduce native account abstraction and create a path toward post-quantum authentication. Together with two companion proposals, the change could also provide protocol-level building blocks for privacy applications.
The “must ship” designation means the two proposals effectively define the upgrade. If either is at risk, the Foundation said Hegotá’s schedule should adjust before the proposal is dropped.
Beyond the two headliners, the Foundation placed 15 proposals in its A-tier, meaning they are expected to ship unless development or testing constraints force cuts. Another eight remain candidates for inclusion, seven were placed below the line but not ruled out, and 28 were declined.
Two proposals remain unranked pending mainnet data from Glamsterdam, the Ethereum upgrade preceding Hegotá that is focused on improving scalability and strengthening the network’s base layer. According to the Foundation, client teams could begin implementing Hegotá in late 2026 following Glamsterdam.
Ethereum’s last major upgrade, Fusaka, went live on Dec. 3, 2025, bringing a series of scaling and usability improvements to the network. Its headline feature, PeerDAS, changed how nodes handle rollup data, reducing the amount each node needs to download and upload while increasing data capacity for Ethereum layer-2 networks.
AI outlook — possibilities, not facts
Client teams begin Hegotá implementation following Glamsterdam.
Likely · Within months

A September 5 proposal to modify Ethereum's EIP-8141 would allow some nodes to accept privacy transactions above a 100,000-gas validation limit, but benchmark data shows required proof verification costs exceed this limit, raising concerns about network-wide support for private transactions like Tornado Cash and RAILGUN.

Harmony blockchain plans to shut down its network on Sunday, citing escalating threats from AI agents and state actors, and proposes migrating its ONE token to Ethereum while launching an AI video initiative called 'The Remix Economy for AI Video'.

Ethereum co-founder Vitalik Buterin dismissed predictions that AI could cause a 50% Bitcoin crash within two years, arguing Bitcoin can withstand security failures not requiring social consensus and that AI-driven exploits are likely fixable. He contrasted with AI-risk commentator Liron Shapiro, who sees a 50% probability of such a crash due to AI undermining confidence in Bitcoin's security. The debate highlights growing concerns over AI's impact on crypto infrastructure, citing real-world examples like AI-assisted attacks on Bitcoin swap provider Boltz and industry-wide defensive initiatives involving Anthropic, Google, Microsoft, and JPMorgan Chase.

ARK Invest and Glassnode published a report on Sept. 1 measuring blockchain capture risk by determining the smallest number of block-production entities needed to cross a protocol-relevant control threshold. The threshold is three entities for Bitcoin and Ethereum, and 19 for Solana. The report highlights differences in network exposure across consensus mechanisms, infrastructure, software diversity, and exit speed, urging institutions to define their failure tolerance before selecting decentralization metrics.

A CryptoSlate analysis shows that in the XRP Ledger lending model, depositor losses from 100,000 tokens of bad debt vary by a factor of 20 depending on whether the debt is in one large loan or ten smaller ones, due to how reserve coverage is calculated per default under fixed cover rates.

Hyperliquid announced a testnet-only extension called HIP-3* that allows independent market deployers to add optional wallet allowlists to their perpetual markets, enabling granular access controls without affecting existing markets or shifting responsibility to the protocol.