Ethereum Research Daily
Ethlabs posted its pqai27 proposal on Ethereum Magicians, arguing that Ethereum should become post-quantum and post-AI hardened in 2027. Hegotá would prioritize Frames and post-quantum accounts, FOCIL would step aside, and Ethlabs is withdrawing quick slots. It gets a short slot on tonight's ACDC #191, and the account-related EIP-8141, EIP-8250, EIP-8151 and EIP-8288 all picked up new revision PRs.
Image: Ethlabs, Ethereum Magicians
Posted on Magicians at 03:51 on 2026-11-12. The core claim is that Ethereum should reach "pqai-ready" in 2027: consensus is post-quantum and every user has a post-quantum migration path for their account, while keeping the capacity Glamsterdam delivers. "pqai-scaled", with post-quantum transactions and blobs at full capacity, is the stretch goal for 2027.
At 03:56 on 2026-11-12, framed as input into Ethereum's governance process.
At 04:43 on 2026-11-12, noting that the post-quantum work also brings new pq-ready accounts built on Frame Transactions.
Comment posted at 06:49 on 2026-11-12.
Opened at 07:12 on 2026-11-12. Only a frame flagged APPROVE_EXECUTION can set sender_approved, so a SENDER frame with no such frame before it always fails, and builders and mempools can drop the transaction without executing anything.
Opened at 07:12 on 2026-11-12. When nonce_seq equals MAX_NONCE_SEQ and every selected key reads zero, the transaction counts as executable rather than a nonce overflow. Replay protection is unchanged.
Opened at 01:20 on 2026-11-12. Under EIP-8151, ecRecover reads the recovered address's code, so an EIP-8141 validation prefix can depend on a third-party account. This PR adds the matching mempool rules.
Opened at 04:49 on 2026-11-12. One user-selected batch program can prove claims for several transactions and signature schemes at once. The PR puts the cost at 30,000 intrinsic proof-share gas plus 3,000 verification gas per claim.
Opened at 07:12 on 2026-11-12, so one leanSPHINCS key carries the same number (currently 0x3) in both EIPs.
Opened at 04:50 on 2026-11-12. Building on EIP-8288, it lets raw signatures go straight to aggregators without first producing a STARK.
Frame Transaction Breakout #8 on 11/17 (pm #2295), and whether pqai27 changes the scope of Frames, since the proposal mentions possible lighter variants.
Opened by LukaszRozmej at 07:12 on 2026-11-12. It adds a code_hash to EIP-7928's AccountChanges, so the BAL names every piece of pre-block state a block reads and a stateless client or prover can build its input from the block, the BAL and a state snapshot without executing the block. The Magicians discussion thread opened at 07:14 on 2026-11-12.
New draft EIP: Code Reads in Block-Level Access Lists Opened by LukaszRozmej at 07:12 on 2026-11-12. It adds a code_hash to EIP-7928's AccountChanges, so the BAL names every piece of pre-block state a block reads and a stateless client or prover can build its input from the block, the BAL and a state snapshot without executing the block. The Magicians discussion thread opened at 07:14 on 2026-11-12.
An opinion post by cskiraly at 18:40 on 2026-11-11. It argues for adjusting capacity to measured resource pressure, the way TCP congestion control does, instead of a fixed gas limit plus a safety margin. Citing mainnet data, it says the median block reaches 95% of reporting nodes 0.38 seconds after it is first seen (0.70 seconds for the 99th percentile block).
At 21:38 on 2026-11-11: private ETH payments with proofs generated in the browser, hiding sender, recipient and amount. The contracts cannot be upgraded or paused, and users can withdraw on their own.
At 21:38 on 2026-11-11: the ceremony recorded 176 contributions, but none were public and one key could contribute repeatedly. The page shows 221 deposits in total, 96 over 30 days and a single active relayer. Know that the anonymity set is small before you use it.
Merged at 04:09 on 2026-11-12, following the spec mismatch that formal verification found earlier this week.
At 22:36 on 2026-11-11: it adds a compute_execution_block_hash check that uses RLP.
Opened at 20:04 on 2026-11-11. It affects how wallets display clear signing formats.
Discussion opened at 01:17 on 2026-11-12.
Discussion opened at 01:03 on 2026-11-12.
At 03:55 on 2026-11-12. The post says each header costs about 626k gas, 361k of it for Equihash.
At 04:25 on 2026-11-12, summing it up as "make Hegotá's headliners PQ and Frames, pause FOCIL, ship PQ in 2027".
At 04:34 on 2026-11-12.
At 05:06 on 2026-11-12.
At 06:07 on 2026-11-12.
At 04:37 on 2026-11-12, a typical community question about whether validator migration is feasible.
At 15:45 on 2026-11-11, with about 1,700 likes (read on 2026-11-12), echoing pqai27's point about faster formal verification. Toni Wahrstätter replies that the hard part is whether the proven spec matches the informal intent At 23:19 on 2026-11-11.
Lightclient shares DeepMind's AI-written Lean proofs verifying the Signal protocol At 15:45 on 2026-11-11, with about 1,700 likes (read on 2026-11-12), echoing pqai27's point about faster formal verification. Toni Wahrstätter replies that the hard part is whether the proven spec matches the informal intent At 23:19 on 2026-11-11.
With 11 posts as of 2026-11-12, focused on account protection at the token level.