Ethereum News
We collect news from the official Ethereum blog, research forums (ethresear.ch, Ethereum Magicians), key Twitter accounts and the community every day and publish it as a digest.
2026-11-13 — A reprioritization proposal to move the quantum-resistance transition forward to 2027 has been submitted to governance
The centerpiece of this issue is the pqai27 position paper Ethlabs posted to Ethereum Magicians. It proposes transitioning Ethereum to post-quantum and, where possible, post-AI resistance by 2027, a timeline that is ahead of the US government's 2030 target and Google's 2029 target. The paper names suspending FOCIL and delaying quick slots as the cost, and fierce debate is expected at the upcoming core developer calls. Meanwhile, the Glamsterdam gas repricing has frozen decisions until the audit results are released on November 16, and formal verification has caught real spec mismatches, making clear that pre-upgrade verification procedures are becoming institutionalized. With the ETH store-of-value narrative now being used as grounds for the security roadmap, this period confirmed a shift in which asset narratives and protocol schedules are being bundled into a single package.
Insights from this issue
- Reprioritization discussion to pull the quantum-resistance transition target forward from 2029 to 2027 has been formalized (ethereum-magicians.org) — After spending a week with core developers at Devcon, Ethlabs submitted the pqai27 position paper as governance input. Its core is a reprioritization that makes quantum resistance and Frames the headliners of Hegotá while suspending FOCIL. Technical readiness is advancing on the same timeline: PQTS Breakout #18 covered operational parameters such as Lean SPHINCS signature lifetime, and the Daisugi testnet is already running with that parameter set. Encrypt The Mempool #12 also forecast that the post-quantum timeline will accelerate significantly, foreshadowing fierce debate at the upcoming ACD calls. Given that a 2027 proposal has now entered a schedule that, through the previous issue, had been oscillating between an end-2029 target and reports of end-2028, whether it is accepted should be watched at the calls from mid-November onward.
- ETH store-of-value positioning is being combined as the justification for the security roadmap (@binji_x) — Following Vitalik Buterin's framing of ETH as a store of value in the previous issue, Ethlabs figures have now linked the same positioning to the justification for accelerating quantum resistance. The pqai27 position paper itself describes the transition as one meant to protect a 'root of trust centered on the programmable store of value ETH,' binji_x called Ethereum a sanctuary for value, and barnabemonnot said that once the quantum-resistance transition is complete, no asset with better properties is visible. Asset narratives have begun to be used as grounds for setting protocol priorities, and a decision structure is forming in which asset safety takes precedence over adding new features. Adding in Kain's retrospective on the EIP-1559 burn meme, the decline of the fee-burn narrative and the rise of the store-of-value narrative are proceeding in parallel.
- Audits and formal verification are becoming institutionalized as preconditions for upgrade decisions (@vdWijden) — For Glamsterdam gas repricing, as Marius van der Wijden stated, both the spec changes and the Hoodi testnet decision have been held pending, with the goal of completing the audit report before ACDT on November 16, and the ACDT #103 agenda includes audit presentations from ChainSecurity and Dedaub, app team contract investigation results, and gas invariants repository results. During the same period, the EF stated that formal verification is becoming the new normal, noting that verity's formal verification of EIP-8282 uncovered a real mismatch for researchers to resolve. The deployed-code compatibility discourse carried over from the previous issue has thus moved beyond warnings and documentation into the decision process itself. Adding gakonst's proposed direction toward a reusable property library, the next thing to watch is whether verification output accumulates into a lasting structure rather than remaining one-off.
Core dev calls
- Encrypt The Mempool #12, November 11 (Forkcast) — In EIP-8184 (Lucid), it was decided to split the key publisher optionality problem out into a separate EIP to reduce cognitive load, and Jannik Luhn published EIP-8491, the key publisher priority list. The blind sandwich paper drew methodological criticism for lacking a null-hypothesis baseline, and the assessment was that its Ethereum-related findings are not meaningful while findings on other chains are more serious. It was also pointed out that the findings could actually be evidence of relay leakage from private mempools, which would strengthen the case for encrypted mempools. consensus-specs#5403 remains stalled, waiting on consensus-layer developer review. It was concluded that including encrypted mempools in I-Star is completely unrealistic and that a realistic target is J-Star at the earliest, with the post-quantum timeline expected to accelerate significantly and fierce debate expected at upcoming ACD calls.
- Post Quantum Transaction Signatures Breakout #18, November 11 (Forkcast) — Whether the Lean SPHINCS key lifetime should be 2^30 or 2^24 remains unresolved. Antonio said that 2^24 (about 16 million signatures) is sufficient, but he prefers 2^30, which is large enough to remove the anxiety of having to count signatures, while accepting that starting at 2^24 is acceptable. Matteo Vena of Project 11 said a large lifetime hurts the performance of constrained hardware such as MPC or hardware signers, and said he would publish benchmarks within days. Amit presented a fixed-point Falcon for Cortex-M4 that removes the floating-point dependency, incurs only a 2-bit security loss, and narrows the performance gap with Dilithium from 4x to 2x. The group reaffirmed its position that Ethereum L1 has settled on hash-based signatures and that Falcon is tracked only for awareness of the L2 and account-abstraction ecosystem. The Daisugi testnet is running with interim parameters and will adopt the sphincs.art parameters starting with the next deployment.
Protocol updates
- Glamsterdam gas repricing audit results to be released at ACDT #103 on November 16 (ethereum-magicians.org) — The ACDT #103 agenda includes the audit status of Glamsterdam EIP-8037 and gas repricing, app team contract investigation results, a Q&A on the audit result disclosure, and gas invariants repository results. The ChainSecurity and Dedaub audits are scheduled to be completed between November 16 and 19. Geth's Marius van der Wijden stated that no decision on spec changes or Hoodi testnet deployment will be made before the audit report, and Ethlabs also recommended not making Glamsterdam decisions before clarity is secured.
- verity discovers a real spec mismatch through formal verification of EIP-8282 (@ethereumfndn) — The EF stated that formal verification is becoming the new normal, noting that while verity was formally verifying EIP-8282, it found a real mismatch that researchers need to resolve. Alongside the crowdsourced formal verification and GKR soundness proof covered in previous issues, this is a case of verification tools beginning to produce real-world results.
Forum proposals
- Ethlabs's pqai27 position paper proposing achieving quantum resistance and AI resistance by 2027 (ethereum-magicians.org) — Ethlabs posted a position paper to Ethereum Magicians proposing that Ethereum transition to quantum resistance, and where possible AI resistance, by 2027. The intended cryptography is hash-function-based with minimal assumptions. Against AI bug hunters, the plan is to prepare through formal verification and client diversity, while also gaining utility such as quantum-resistant accounts based on Frame transactions. The cost is suspending FOCIL and delaying quick slots, with the view that the consensus-layer change (FOCIL) could be deferred from Hegotá while the execution-layer change (Frames) could still ship as planned. ryanberckmans agreed, calling quantum resistance Ethereum's most important technical milestone, while decentrek summarized it as a request to move the current target forward by two years.
- EIP-8491 LUCID Key Publisher Priority Lists, which determines priority ordering for sealed transactions (ethereum-magicians.org) — In this draft published by Jannik Luhn, validators subscribe to an on-chain-managed list of key publisher addresses via BLS-signed messages, and an address becomes priority once it receives support exceeding 50% of active validators' effective balance. Priority publishers' sealed transaction commitments must be ordered first in the block; violations invalidate the payload, and publishers are forced to sign a key-disclosure message. The call also noted a limitation: removing max_preceding_commitments in EIP-8184 addresses probabilistic front-running but does not address spam risk from exploiting proposer delay.
- EIP-8496, which records code reads in the block-level access list (ethereum-magicians.org) — Nethermind's Lukasz Rozmej posted a draft adding a code_hash field to each AccountChanges entry of the EIP-7928 block-level access list (BAL). If a block actually loads the bytecode before execution, it records the Keccak-256 hash; otherwise it records an empty value. This changes the RLP format and block_access_list_hash, requiring a hard fork. Together with EIP-7709, the goal is to assemble stateless execution and proof inputs using only the block, the BAL, and a state snapshot.
- eth/73, a devp2p extension for propagating the EIP-8288 proof wrapper (ethresear.ch) — A protocol draft has emerged that extends eth/72 to propagate the EIP-8288 proof wrapper between peers. It defines new messages NewProofWrapperHashes (up to 64), GetProofWrappers (up to 32), and ProofWrappers (soft limit 2MiB), along with compact serving and a raw-witness wrapper mode. Following the EIP-8288 OP Stack PoC covered in the previous issue (a 96% reduction in blob usage), the networking layer is now being filled in.
- An Elastic Ethereum vision piece proposing that throughput be adjusted according to demand and capacity (ethresear.ch) — cskiraly posted an opinion piece on ethresear.ch. It points out that while Ethereum's slot time, gas, and size limits are preset and adjusted slowly, actual network and node capacity change quickly, and there is no protocol-wide closed-loop control that reduces load when the system falls behind. Drawing an analogy to TCP, it presents the backlog of two buffers, the mempool and the 'available but not yet finalized' chain, as the missing control signal, and proposes two design directions: setpoint control of how far block building runs ahead of finality, and lowering reorg cost through a selective-ack voting window and recursive aggregate proofs.
- Seam, which verifies Zcash headers on Base without a bridge (ethresear.ch) — A permissionless relay on Base verifies Zcash consensus rules, including Equihash, using the EIP-152 BLAKE2b precompile, a design that lets a contract read the Zcash best chain without trusting a bridge or a committee. It is estimated to cost about 626,000 gas per header on average, or about 0.01 ETH per day when following 25-second blocks, and on top of that it uses a RISC Zero zkVM proof to prove that a shielded payment meets a condition without exposing a transaction ID or address. It is at the testnet stage, unaudited, and its security bound is an economic approach that requires cumulative work.
Twitter debates
- lightclients and nero_eth split over whether reading only the spec is enough in formal verification (@lightclients) — When lightclients summarized the benefit of formal verification as 'you don't need to read the proof, just the spec,' nero_eth initially agreed but then asked, 'what if you don't understand the spec?' GiulioRebuffo replied that since the spec is far simpler than the implementation, not knowing the spec means not knowing the implementation either. nero_eth countered that starting from the spec is itself too late in the pipeline, that the informal spec needs to be sorted out first, and that the real hard problem is trusting that the proven spec matches the informal spec. It went as far as lightclients suspecting the original post was bait.
- Researchers and app developers continued to weigh in on the charging design of the Glamsterdam gas repricing (@barnabemonnot) — barnabemonnot recalled that researchers have for years received requests from part of the community to raise gas significantly regardless of what happens to small nodes. When Leo_Glisic asked why bytecode read costs are kept as a fixed per-contract charge rather than per-byte like writes, barnabemonnot replied that once code chunking arrives together with the Partitioned Binary Tree, reads will also be charged per byte, and that reads can be less strict than writes because they do not impose a permanent cost. On the other side, Austin Griffith said in an interview that some operations have become more expensive, leaving app developers angry. The Versioned Opcodes remediation covered in the previous issue has now been assigned EIP-8488 and advanced to a formal proposal.
- Positions split after reports of litigation over the EF Mandate's pledge clause and a firing (@laurashin) — According to Laura Shin's reporting, former EF staffer Kayleen Fulton's lawsuit claims that 79 staff voted to remove the so-called seppuku pledge from the Mandate. The published Mandate does not contain that word in a search, but it does contain language to the effect that if Ethereum fails, the Foundation should take responsibility itself and step down. TrustlessState said that if there was a firing depending on whether the document was signed, it cannot be dismissed as a joke, and 0xzak and Greg Markou argued that such a document could not have been issued without executive approval. Kain, on the other hand, criticized those who make removing the pledge clause their top priority as acting in bad faith to stir controversy, and hasufl defended the outcome, saying the Mandate succeeded in driving the necessary downsizing and increasing ecosystem diversity.
- peter_szilagyi and a rebuttal clashed over IoT vulnerability disclosure process and impact assessment (@peter_szilagyi) — Geth's peter_szilagyi revealed that he found four or more vulnerabilities, including three remote takeovers, in a home automation device he uses, but that the $750 million-revenue Swiss vendor has no security disclosure channel and did not respond to email. The anonymous account fbi__open__up countered that critical infrastructure would not expose such devices to the internet, calling the impact overstated, and peter_szilagyi rebutted that the quality of consumer products itself is the problem. He later submitted the report through CERT/CC's VINCE channel but, objecting to the criminal liability notice in the first line, suggested he would abandon the report, and recommended immediately disconnecting network-connected smart devices. His claim that he used a China-hosted model because Western AI models refused to even examine his own device echoes the local-inference-switch declaration from the previous issue.
Notable remarks
- gakonst: 'The value of formal verification lies in a reusable library of properties' (@gakonst) — gakonst argued that the bull case for formal verification is not handing over millions of lines of Lean proofs and saying a model passed them, but providing a reusable library of properties and specs that can be imported into code. The remark offers direction amid the trend of the EF framing formal verification as the new standard, citing verity's EIP-8282 verification.
- Kain: 'The intent of EIP-1559 was not ETH burning' (@laurashin) — In an Unchained interview, Kain recalled that burning was merely a quirky mechanism design choice, not the goal of EIP-1559, and that as ultrasound money became too effective a meme, the discourse degenerated into how to burn more. This echoes peter_szilagyi's criticism of the issuance-reduction debate and Vitalik Buterin's store-of-value declaration from the previous issue.
- Greg Markou: 'It is hard to know in advance which change will break which contract' (@laurashin) — Sprinter co-founder Greg Markou called Ethereum a monolithic beast and said invariants are hard to pin down, recounting how, while maintaining web3.js in the past, he unknowingly broke almost every app after EIP-1559. Even with testnets in place, it is not easy to assert with confidence which contracts a given change will break.
- binji_x: 'ETH is a store of value and Ethereum is a sanctuary for value' (@binji_x) — binji_x described Ethereum's ambition as accepting and protecting all value that cannot be entrusted to a single company, government, or server, characterizing ETH as an asset that inherits Ethereum's properties one-to-one. Coming out the same week as the pqai27 announcement, this messaging shows coordinated positioning around Ethlabs that bundles the quantum-resistance roadmap and the store-of-value narrative into one package.
Ecosystem & security
- Claim that Ledger devices send telemetry to a server after signing (@laurashin) — In an Unchained interview, Greg Markou claimed that after a signing interaction on a Ledger device, a POST request containing metadata is sent to a server, and that even if some metadata is removed in v2, addresses and users can still be correlated. He said telemetry is sent even when signing fails or an error occurs, and pointed out that sending anything after signing is entirely unnecessary. This extends the clear signing flaw discussion raised earlier by 0xMidnight.
- On-chain security product Phylax launches (@barnabemonnot) — odysseas_eth's Phylax has launched, and barnabemonnot called it a wholly new addition to the arsenal of defensive tools, praising it as a product with clear purpose and great necessity. Aimed at improving application security to make insurability more attainable, it connects with pcaversaccio's concern raised in the previous issue that Ethereum security does not get funded on its own.
- LlamaRisk proposal to restructure Aave V3 Core around blue-chip collateral (@StaniKulechov) — LlamaRisk has put forward a proposal to reposition Aave's largest market, V3 Core, around highly liquid blue-chip collateral, removing restaking assets, leveraged wrapping strategies, and issuers lacking operational and security resources, and former V3 builder ameensol backed it. Separately, Aave V3 has upgraded 15 markets to V3.8 following DAO approval, with no action required from users or integrators.
- Aave introduces Chainlink CRE-based dynamic borrow rate limits (@StaniKulechov) — Aave is introducing dynamic borrow rate limits that control the real-time rate of debt growth per asset, going beyond static borrow caps. It runs on Chainlink CRE, developed by LlamaRisk and reviewed by Aave Labs. This turns the borrow-rate-limit preparations reported in the previous issue, made in response to the Binance hack funds, into an official announcement.
- Agent identity verification added to the Machine Payments Protocol (@gakonst) — gakonst announced that identity support has been added to the Machine Payments Protocol, saying that while an agent with a wallet can access any service, services do not want to accept just any agent. A round trip was added to the 402 response resubmission flow so that an API can require proof of email ownership, wallet lookup, or third-party reputation checks, and following integrations with Boundless (inference) and e2b (sandboxing), the team is now recruiting model routers and cloud partners.
Market briefing
Other news
- RPC Standards #38 call to be held November 16 (ethereum-magicians.org) — The agenda includes rejecting pre-fork-inactive transaction fields, eth_call ignoring the type and rejecting maxFeePerBlobGas without a blob hash, eth_createAccessList returning the access list, and eth_simulateV1 fixtures.
- Five new ERC drafts this week, including a token metadata schema (ethereum-magicians.org) — ERC-8495 (token metadata JSON schema), ERC-8446 (cross-chain client gateway for CCIP-Read), ERC-8493 (on-chain URI scheme), ERC-8494 (institutionally managed user payment ledger), and ERC-8492 (task escrow settlement deduction) were newly submitted.
2026-11-10 — Compatibility with deployed code became a precondition for activating Glamsterdam
The Glamsterdam gas repricing debate is the focus of this issue. ryanberckmans, who had argued for prompt activation, publicly withdrew that position within a day, and the weight shifted toward the view that an upgrade that breaks deployed contracts cannot go ahead as is. Two compatibility-preserving repricing EIPs were posted as remedies, and discussion began on where to codify the invariants that must not be broken. Separately, Vitalik Buterin issued an official position that it is reasonable to view ETH as a store of value rather than a fee-burning token, and reactions to it were split. Over these three days, both the technical debate and the asset discourse redefined what Ethereum guarantees and what it aims for.
Insights from this issue
- Compatibility with deployed code is hardening into a requirement for hard fork activation (@ryanberckmans) — In the previous issue, the EIP-8037 and EIP-8038 repricing problem was at the stage of an official warning. Over these three days it grew into the issue that decides whether the fork activates. ryanberckmans, who had pushed to force activation, publicly withdrew within a day, saying 'the problem is more serious than it is generally understood', and ultimately stated that 'it cannot be activated before it is resolved'. In the same period, compatibility-preserving remedy EIPs were posted one after another, including Versioned Opcodes, which keeps existing contracts on the old gas schedule, and Double-Pricing, which separates execution from billing. EF's fradamt proposed making contract impact scanning a standard upgrade practice. With discussion even starting on codifying the invariants that must not break in the EIP process, the next things to watch are adjustments to the Glamsterdam schedule and whether the remedy EIPs are adopted.
- ETH's store-of-value positioning has become the founder's official position (@VitalikButerin) — The previous issue noted that the discussion of ETH's asset identity was converging on programmable store of value. This time Vitalik Buterin himself published a long position piece arguing that 'it is reasonable to view ETH as a store of value (SoV) rather than a fee-burning token'. It is a three-part argument: fee-based valuation was weak to begin with, maximizing fees requires compromising decentralization, and optimizing for SoV aligns with the neutrality principle. peter_szilagyi criticized the entire issuance debate as 'avoiding the fundamental problem that there is nothing to do onchain', and BitMine decided to stop accumulating at 5% of supply at the request of Vitalik and Justin Drake. Both touch the same question of what ETH as an asset should be managed for. It is worth watching whether the SoV frame gets cited as a decision criterion in the issuance and fee policy debates.
- App developers have begun demanding a role in upgrade decisions (@ryanberckmans) — Wildcat Labs' public protest post, the observation that 'smart contract developers are underrepresented in upgrade planning', the retort to 'participate directly in the ACD calls', ryanberckmans' proposal to widen communication lines between app developers and core developers, and rakitadragan's argument for EF board accountability: the second front of the repricing episode was not technical but about who takes part in decisions. barnabemonnot proposed separating parameters from pricing mechanisms and documenting what can change and what must be preserved, and a repository for app developers to self-diagnose impact was also released. The key question is whether the process moves past venting anger to institutionalizing participation and documentation.
Protocol updates
- Versioned Opcodes, a repricing remedy that permanently applies the old gas schedule to existing contracts (ethresear.ch) — A proposal titled 'Forward-Compatible Gas Repricing via Versioned Opcodes' was posted. It permanently applies the current gas schedule to already deployed contracts (grandfathering) and has only newly deployed code use V2 opcodes that follow the new costs of EIP-8037 and EIP-8038. V2 code forbids gas introspection opcodes such as GAS and GASLIMIT, and introduces a code_version=2 marker and DELEGATECALL_V2. It is a compatibility-preserving remedy that arrived in the middle of the repricing debate, and the burden on clients of maintaining two gas models side by side is expected to be a key point of evaluation.
- Double-Pricing, a proposal to separate execution gas from billing gas (ethresear.ch) — A proposal titled 'Introduce Double-Pricing' was posted. It separates execution, run on credits from the legacy schedule, from billing, run on the new schedule, so that behavior observable in the EVM, such as gasleft(), stays identical to before. When billing runs out, the whole transaction is rolled back with BILLING_OUT_OF_GAS and charged in full. The approach aims to honor the gas assumptions of existing contracts while still obtaining the economic effect of higher state costs.
- Builder DoS problem raised against repricing alternatives, and a mitigation (@nero_eth) — On the idea of 'charge as before, then settle under the new rules at the end of the transaction', EF's nero_eth pointed out that block builders would be exposed to DoS because block fit cannot be checked against the transaction gas limit. As mitigations, he cited giving transactions two gas limits and the fact that the 16.7M transaction gas limit acts as a buffer, and as an alternative he also proposed adding state gas to SSTORE and extending EIP-2930 access lists as an escape hatch for prepaying. He also noted a limitation: because state access is determined at runtime, access lists can only serve as an escape hatch, not a default.
- Correction: the EIP-8261 gas limit schedule is an optional EIP, not an informational one (@BarnabasBusa) — After it was pointed out that EIP-8261 is informational, EF DevOps' BarnabasBusa corrected that it is an optional EIP. It works by setting client defaults, and he said that all clients except Nimbus currently implement it, so the gas limit schedule will actually take effect at the fork. It shows by what mechanism the gas limit increase (from 60M to 200M), which is the premise of the repricing debate, gets enforced.
Forum proposals
- Mempool DoS benchmark of the EIP-8141 verification budget: the current 100k is not enough for post-quantum signatures (ethresear.ch) — Nethermind researchers (including Stefano De Angelis) benchmarked the MAX_VERIFY_GAS cap for frame transactions (EIP-8141). The current 100k gas cannot fit a Groth16 privacy transfer (about 212k) or post-quantum signature verification (about 276k), and adding a client protection that pauses frame verification during block import could raise it to about 300k. However, a limit was also measured: under an attack of 100 transactions per second, the acceptance rate of honest frame transactions falls to a range of 12 to 27%. The frame transaction format that has been converging since September has entered a stage of numerically verifying whether it can handle real post-quantum and privacy use cases.
- leanVM post-quantum signature aggregation: measurements show the folding approach is not practical (ethresear.ch) — A negative result was published showing that directly folding leanXMSS signatures costs about 7.1 million R1CS constraints per signature, roughly 450 times the cost of running it in leanVM, so it is not practical. The practical lesson is in setting the bounds of the recursion parameter mu: for 1,500 signatures, changing the aggregation configuration from 250 groups of 6 to 300 groups of 5 cut the deadline from 3.11 seconds to 2.24 seconds. It is also significant as a research practice of sharing dead ends to reduce duplicated exploration by other teams.
- Machine-checked soundness proof of the GKR protocol and sparse prover performance improvements (ethresear.ch) — A post was published reporting the first machine-checked soundness bound for general layered-circuit GKR, done in Isabelle/HOL. Using a degree-4 extension of the KoalaBear 31-bit prime field, it secures a security denominator of about 2^124. The sparse prover 1.2, released alongside it, raises throughput by about 2.4 to 2.7 times while keeping proofs byte-for-byte identical. It is work on building a trust foundation for proof systems, tied to the crowdsourced formal verification effort that has continued since September.
- ERC draft Protected Accounts: an opt-in protection mode built into ERC-20 accounts (Ethereum Magicians) — A draft ERC-20 extension was posted in which turning on a per-account protected flag lets small amounts transfer immediately while large transfers are delayed so that a backup wallet can cancel them. In a physical coercion situation (a wrench attack), a stress action can move the entire balance to the backup wallet and lock it. It belongs to the same line as last week's proposals to put spending limits and recovery at the asset level rather than the account level, and the trend of building security features into the token standard itself continues.
- Concept proposal to move lattice operations into an HSM to avoid the post-quantum bandwidth problem (ethresear.ch) — A concept-stage post was published proposing to avoid the validator bandwidth problem by moving lattice-based operations inside a hardware security module (HSM) and combining it with an anti-tamper identity based on the quantum no-cloning principle. It is more a statement of the problem than an implementation.
Twitter debates
- ryanberckmans publicly withdrew his position on activating the Glamsterdam gas repricing (@ryanberckmans) — ryanberckmans argued for prompt activation, saying 'Ethereum's promise is not gas prices but access to a neutral state transition function'. gakonst (Paradigm) rebutted that 'the current Glamsterdam is not CROPS and core developers did not analyze it properly for 1.5 years', and lex_node said 'immutable code continuing to work is the reason Ethereum has been advertised to exist for years'. Security researcher pcaversaccio stated flatly that cancelling the fork in its current form would itself be a success. Within a day ryanberckmans withdrew, saying 'the problem is more serious than it is generally understood', and the next day gave his final position that 'it cannot be activated before it is resolved'. Meanwhile, Geth's vdWijden drew a line against the opt-out in which only existing contracts use the old gas model, citing the need to maintain two EVMs and the risk of consensus bugs. Reth's rakitadragan raised accountability by pointing to ryanberckmans' position on the EF board.
- TrustlessState, barnabemonnot and others split over how to assess Vitalik's store-of-value declaration (@TrustlessState) — Bankless' David Hoffman (TrustlessState) assessed that 'it is true that wanting SoV makes sense since ETH does not work as a fee-generating asset, but ETH is not a good SoV, and a declaration at this point comes from weakness, not strength'. EF's barnabemonnot asked in return whether 'it would have been better not to say anything at all', and ryanberckmans strongly mocked the claim that aligning the community around SoV is a sign of weakness. To those arguing for a return to PoW and removing programmability, Hoffman also drew a line, saying 'that makes Ethereum into something that is not Ethereum'. Separately, lex_node laid out the Zcash camp's argument that 'a smart contract L1 is a bad SoV', then countered that an SoV that allows exchange, lending and management without leaving its native environment is better.
- The camps split over the state tree migration method, online versus self-migration (@donnoh_eth) — gakonst and others criticized the offline (self-migration) approach as far more complex and fragile than the online approach and said coordination would be a nightmare, and ipsilon's chfast took issue with the manner of the debate itself, saying it is trying to win on social media with memes. On the other side, empirical results were presented: four clients succeeded in self-migrating mainnet to a Partitioned Binary Tree in a 16GB RAM and 2TB disk environment, and a consensus layer EIP for shadow root visibility is in progress (speaker attribution is uncertain in the collection record and needs checking against the original). CPerezz19 rebutted that the very fact that additional EIPs and code surface are needed proves the superiority of the online approach. The restructuring of the state design for the I* fork, foreshadowed in the previous issue, has moved on to a fight over the concrete migration method.
- A proposal to defer state mitigation measures drew a counterargument based on XEN risk (@donnoh_eth) — donnoh_eth proposed shipping only the scaling-related changes first in Glamsterdam and deferring the state bloat mitigations, arguing that the supply-induced demand that justifies the mitigations may never materialize. A counterargument was that the XEN token, which is mined simply by writing to state, is already the largest state consumer on Ethereum and a constant top gas user, so any scaling change would immediately lead to an uncomfortable level of state growth. lex_node summed up that Ethereum has reached the stage of a chronic patient, where interventions conflict and improving one problem worsens another.
- barnabemonnot and nero_eth proposed different places to codify the invariants that must not be broken (@barnabemonnot) — barnabemonnot of the EF Robust Incentives Group divided invariants into parameters such as gas costs and block time, and pricing mechanisms. They assumed the former will change but said the latter must be updated non-destructively. EIP-8037 is described as a case where a mechanism update caused an unexpected breaking change and further delay. For where to codify them, one proposal was to run an informational EIP through the PFI, CFI, SFI pipeline so that the process, not a specific actor, decides. nero_eth of the EF disagreed, saying they belong in each EIP's backwards compatibility section or in the fork meta EIP, and that informational EIPs are already overloaded. Testimony also emerged that app developers have used guardrails based on the expected SSTORE cost as common practice for years, showing that implicit norms have actually accumulated.
- Wildcat Labs publicly protested protocol changes made without asking builders (@ben_a_adams) — Wildcat Labs published a post titled "Don't change things without asking the people building on Ethereum" that lists five reasons for fatigue (the collection record is unclear on which author handle it belongs to). One of the points is that geth's default 50M RPC cap makes gas estimation impossible for the combined deployment of Wildcat V2 markets. mysticryuujin objected, asking where today there is a public RPC that caps below 500M. The thread also included an exchange saying that EIP-8298 SETCODEFROM, which cuts factory deployment cost to about one tenth, is also needed for quantum resistance and will likely land in a fork after Hegotá.
- A trilemma formulation emerged on the limits of ETH-overcollateralized stablecoins (@hasufl) — binji_x asked about the potential of ETH-overcollateralized stablecoins and why DAI gave up on ETH as its sole collateral. A reply, presumed to be from hasufl, said an overcollateralized stablecoin needs natural long demand for the volatile collateral, so supply cannot expand in a bear market. It then presented a trilemma in which only two of three properties can be chosen: crypto collateral and immutability, solvency, and scalability. By this classification, the old DAI and LUSD sacrificed scalability, while USDT and USDC sacrificed crypto collateral. donnoh_eth pointed out that Liquity's BOLD also depends entirely on a Chainlink oracle and is not as robust as it is rated.
Notable remarks
- Vitalik Buterin: "It is more reasonable to view ETH as a store of value than as a fee-burning token" (@VitalikButerin) — Vitalik Buterin posted a long thread with three arguments. The fee-burning token logic, which values market cap as the present value of burned fees, was not as strong as originally thought. Justifying a $270 billion market cap through fee maximization would require acting like a global-scale fee-extraction company, which would compromise decentralization and capture resistance. Optimizing for store of value (SoV), by contrast, aligns with decentralization principles. The convergence toward programmable store of value noted in the last issue is now effectively confirmed as the founder's official position.
- peter_szilagyi: "The issuance reduction debate dodges the problem that there is nothing to do onchain" (@peter_szilagyi) — peter_szilagyi, who formerly led Geth, criticized the entire discussion on reducing issuance. He argued that staking concentration is only a symptom and the root cause is that onchain activity itself is risky, that is, "the problem is not that staked ETH creates no activity, but that active ETH is at risk." He wrote that reducing issuance would only make people choose between the casino and exit, so rather than touching monetary policy, people should build things that are useful in everyday life.
- dankrad: "The standard of never breaking existing contracts does not work" (@dankrad) — dankrad, who has moved to Tempo, wrote on the repricing debate that a principle of never breaking existing contracts cannot work, because it would make both gas schedule changes and the retirement of signature schemes, which is essential for quantum resistance, impossible. Instead, he proposed clarifying which kinds of contracts will not be broken, and said contracts that depend on the gas schedule or state commitments should be upgradeable, or users should be aware that they must exit in certain situations.
- fradamt: "Contract impact scanning should become standard practice for upgrades" (@fradamt) — fradamt of the EF posted a long analysis of repricing impact. Because the gas limit rises together, a gas increase on state operations does not mean a proportional cost increase, and items that actually become more expensive, such as new state writes, will be targeted by future scaling work. However, fradamt admitted that contract impact research should have been done earlier and in more depth, and noted that a few agents and the Sourcify database were enough to find the affected contracts. fradamt therefore proposed making contract scanning against the latest spec a standard practice for protocol upgrades.
- Barnabé Monnot: "Slow is not a virtue" (@barnabemonnot) — Barnabé Monnot agreed with the view that fast slots should be treated with caution but still defended shortening slots. He argued that shorter slots are positively correlated with censorship resistance, and even with builder market health, as long as a decentralized network supports them. Both security and utility should be targeted, and Ethereum is not yet anywhere near the efficiency frontier.
Ecosystem & security
- Demo released of an agent paying for inference costs in real time (@gakonst) — gakonst of Paradigm shared a demo in which an AI agent pays for inference costs in real time and tops up the session when it hits a limit. He also raised demand for MPP support alongside x402, and for services such as Postgres where agents pay the fees. The flood of agent credit and payment ERC drafts seen last week is moving into the stage of real product demos.
- Agents on Base connected to trading on Coinbase (@base) — Base announced Bankr's Coinbase connection, saying that agents built on Base trade on Coinbase. Coinbase Wallet's AI agent feature is also slated for release within the year, so points of contact for agents are opening on both the exchange and the wallet side.
Market briefing
- Tom Lee's BitMine to stop accumulating at 5% of ETH supply (@ETH_MAXER (relayed from the interview)) — In an interview dated November 8 relayed by ETH_MAXER, Tom Lee said about 40,000 ETH remain to reach the holding target of 5% of supply. At the request of Vitalik Buterin and Justin Drake, who do not want a single entity to hold too much ETH in concentration, he said BitMine will not buy beyond 5% for the time being. Staking income exceeds $400 million a year (about 3%), and he named share buybacks as the top priority going forward. Since holdings were 6.06 million ETH, about 4.9% of supply, per last week's 8-K, reaching 5% within the year is likely.
- Outlook for the US House market structure bill CLARITY to stall, and Democratic alternatives (@laurashin (Unchained)) — On the Unchained broadcast, Justin Slaughter said after the midterms that "CLARITY looks dead." The alternatives House Democrats are floating include stronger ethics rules for public officials, tighter DeFi regulation, and moving jurisdiction from the CFTC to the SEC. The Democratic move to shift jurisdiction to the SEC covered in the last issue is taking concrete shape as a change in the legislative landscape.
- XLayer's tokenized stocks xStocks reach $165 million in trading in one month (@Optimism) — Optimism reported that xStocks, tokenized stocks on the OP Stack chain XLayer, recorded $165 million one month after launch. It continues the expansion of tokenized stock infrastructure covered in the last issue, including Uniswap passing $20 billion in cumulative volume.
Other news
- Report of a December 2028 target for the quantum-resistance and post-AI transition (@EtherealnewsHQ) — The weekly newsletter Ethereal news #46 reported December 2028 as the target date for the quantum-resistance and post-AI transition. This differs from the previously known December 2029 target for a quantum-resistant L1, so the original source needs to be checked.
- Two ERC drafts on agent self-management and information flow (Ethereum Magicians) — ERC-8490, in which an AI agent records its maintenance history in a versioned, attributable registry, and ERC-8489, which defines a content-addressed information layer between agents, were submitted by the same author. Agent-related standard drafts keep increasing.
2026-11-07 — Urgent client bug patches overlapped with the Glamsterdam invariant warning, making checks of deployed code a shared task
Both main threads of this issue point at code that is already deployed. A Reth state root bug was found through fuzzing and led to an urgent patch across the ecosystem, and during the same period the EF officially warned that Glamsterdam's gas repricing (EIP-8037 and 8038) breaks invariants of some deployed contracts. As protocols and clients change quickly, who checks the safety of deployed code, and how, has emerged as a standing issue. Meanwhile, as Devcon 8 in Mumbai closed, the debate over ETH's asset identity, "is ETH money or a store of value," became a public discourse that even EF figures take part in, and tokenized stocks have entered a phase in which infrastructure expansion and collective opposition from issuing companies are growing at the same time.
Insights from this issue
- The risk of protocol changes breaking deployed code has started to be managed through a standing check system (EF Blog) — An edge case in Reth's legacy state root code, where a crafted transaction could produce an incorrect state root, was found through fuzzing by the Tempo team, and patches in Reth v2.10.0, OP Reth v2.8.1 and Base v1.5.3 were shipped immediately. At almost the same time, the EF blog announced that Glamsterdam's EIP-8037 and EIP-8038 gas repricing breaks invariants that some projects depend on, and Geth's vdWijden published a self-check repository collecting the invariants that break after the fork. Following on from the default path for handling the gas repricing in late October being set as "self-fixing by affected contracts," tools are being put in place one by one: fuzzing, an invariant repository, and self-checks by deployers. The next thing to watch is how much deployers' actual participation in checks rises before Glamsterdam activates.
- The debate over ETH's asset identity is converging on programmable store of value (@lex_node) — After word spread that Vitalik Buterin said "ETH is money" at Devcon, binji_x of the EF agreed that ETH has become a legitimate store of value (SoV) but drew a line, saying it is not yet money and that overcollateralized ETH stablecoins could fill that gap. lex_node commented that the ecosystem is aligning toward defining and marketing ETH as a "programmable store of value," and that ETFs and institutions will welcome this, and EF researcher Julian Ma also previewed an announcement about ETH as a store-of-value asset for capital allocators. It is in the same vein that when Stani Kulechov invoked Bitcoin during the Paraguay banking crisis, lex_node countered that ETH, which supports borrowing, swaps and even privacy, is a better fit. The issue on which lex_node and Tim Beiko split last month has turned into a shared narrative within a month, one that even EF figures take part in.
- Infrastructure expansion in tokenized finance and resistance from issuing companies are growing together (@ETH_MAXER) — Uniswap connected Groma's real estate issuance and redemption directly to the protocol through its Permissioned Pools Hook, which puts permissioned assets on the AMM, and cumulative tokenized stock trading volume has passed $20 billion. Aave V4's expanded collateral on the Base market, Arcus's tokenized stock margin, and Amundi's tokenization of core European equities show that tokenized assets keep being absorbed into financial structures as collateral and margin. Meanwhile, according to ETH_MAXER, more companies slated to trade on the OKX and ICE tokenized securities exchange have expressed opposition, bringing the count of objecting companies to 67, and no reasons were disclosed. This is up from the 50 opt-outs covered in the previous issue. As the legitimacy dispute that began with AMC's resistance in September spreads across listed companies, and with Democratic moves to strengthen SEC jurisdiction on top of it, the institutional path for tokenized stocks may be redrawn.
Protocol updates
- Reth state root bug discovered and emergency patch across the ecosystem (@gakonst) — An edge case was found in Reth's legacy state root code in which, with certain smart contract storage layouts, a crafted transaction could produce an incorrect state root. It was found through fuzzing by the Tempo team. Block builders on low-spec hardware and OP Stack chain setups were exposed, and patches were released as Reth v2.10.0, OP Reth v2.8.1, and Base v1.5.3. gakonst explained that L2 withdrawal systems are not affected and that bridges are unaffected because they use receipt roots or multisigs rather than the state root, and apologized, saying 'Reth is core infrastructure for the entire EVM ecosystem.' Base patched all of its own block builders on November 4, the day of discovery, and LayerZero temporarily paused some cross-chain routes as a precaution.
- Official warning on deployed contract invariants broken by Glamsterdam's EIP-8037 and EIP-8038 (@vdWijden) — The EF blog announced that Chainsecurity's research found projects that become vulnerable after Glamsterdam. It says that EIP-8037 and EIP-8038, which readjust the gas cost of certain operations and the way state gas is charged, break invariants that some deployed contracts depend on. Geth's vdWijden published a verification repository collecting the invariants that break after the fork and asked deployers to run self-checks, and the EF's ralexstokes and others helped spread the word. This continues from nero_eth's October request to review EIP-8037, and managing the side effects of repricing has now moved into official channels.
- EF publishes fork-by-fork plan toward Lean Ethereum (@ethereumfndn) — The EF published a plan that breaks the Strawmap's '2029 Lean Ethereum' goal down by fork. The schedule is that the H* fork is the last fork without lean features, the I* fork introduces leanVM and lean transactions, the J* fork adds lean attestation, DA sampling, and PCS, and the L* fork completes Lean Ethereum. The Hegotá unified tierlist and the December 2029 goal for a quantum-resistant L1, both covered in the previous issue, have now been made concrete as a fork-level execution plan.
- Preview of the I* fork's state structure overhaul to a Partitioned Binary Tree (@ethereumfndn) — The EF said it will overhaul the state structure into a Partitioned Binary Tree in the I* fork without downtime. The work makes state proofs small enough for wallets to verify directly rather than trust a third party, and opens the path to state expiry, where nodes discard unused data. The discussion of the physical storage design for EIP-8297 binary tree state, covered in October, has now been placed on the fork schedule.
- Hegotá's assertions: guaranteeing transaction effects before signing (@ethereumfndn) — The EF said that user interviews repeatedly produced the answer 'I'm not confident when signing a transaction,' and explained that if assertions are included in Hegotá, wallets will show the exact effects of a transaction and provide a guarantee that it reverts if the outcome differs from the intent. This approach solves, as a protocol feature, the same problem addressed in the October discussion of STVA (a verification standard for preventing blind signing).
- Base Denim mainnet launch: 200ms native blocks and B20 improvements (@base) — Base deployed the Denim upgrade to mainnet. It replaces the Flashblocks preview with 5 full blocks per second (each with its own block number, hash, state root, and receipts), cutting block time from 2 seconds to 200ms, and BaseTime lets contracts handle time in milliseconds (block.timestamp stays in seconds). The B20 token standard was improved to be friendlier to regulated assets, for example by letting issuers force all transfers to go through an approving party such as a transfer agent. The next upgrade was announced as Everest.
- FCR support spreads to RPC providers, removing the adoption blocker for L2s and bridges (@_julianma) — Quicknode has started supporting FCR for all users. The EF's Julian Ma assessed that many L2s and bridges had been blocked by RPC providers' lack of FCR support and that this bottleneck has been cleared, and Barnabé Monnot also said that RPC onboarding unblocks, in a chain reaction, the systems built on top of that signal. The 'move to change bridge confirmation criteria to validator attestations' covered in the previous issue has now reached the infrastructure stage.
- Ethereum Economic Zone documentation and three demos released (@barnabemonnot) — The EF's Barnabé Monnot released the EEZ (Ethereum Economic Zone) documentation and three demos: a CoW order that pulls in L1 liquidity, a case that uses L1 as the root of trust for Safe owner changes, and a flow that approves L1 transfers from a rollup. He explained, 'Increasing L1 utility deepens the L1's function as the center of the Ethereum economy.'
- Frame Transaction Breakout #8 announced for November 17 (Ethereum Magicians) — The 8th breakout on frame transactions (the EIP-8141 family) will be held on November 17 at 14:00 UTC. The agenda covers devnet status and the next devnet plan, spec changes, the test release timeline, client team updates, and feedback from wallet, tooling, and app developers.
Forum proposals
- EIP-8484: SLOTDURATION opcode to read slot duration (Ethereum Magicians) — OpenZeppelin's Hadrien Croubois (Amxx) proposed a new opcode, SLOTDURATION (0x4c, 2 gas), which pushes the current slot's duration onto the stack as a uint64 in milliseconds. It adds a slotDurationMs field to the block header after EIP-7843's slotNumber and requires a new version of the Engine API. The aim is to stop contracts from hardcoding 12 seconds and to prepare for future slot time changes such as EIP-8198. Whether it should return only the current value or also support looking up past slot values (considering Lido-style contracts) is an open question.
- Walraswap: an MEV-resistant DEX design that settles at exchange economy equilibrium prices (ethresear.ch) — A DEX design was posted on ethresear.ch that settles in batches at the exchange economy equilibrium of n tokens (a single clearing price including all orders). It showed that, even with external AMM swaps included, a clearing price exists at which the batch ends with non-negative surplus in every token, formalized the equilibrium computation as a convex optimization, and proved that even when the equilibrium price is not unique, the net trade volume is unique. It addresses solver order censorship with governance in which a DAO elects solvers and order flow is made public, deterring censorship as a repeated game.
- ERC-8486: Flash Loan Proxy standard using transient storage (Ethereum Magicians) — A draft proxy standard was posted that temporarily caches the program and calldata in transient storage and forwards them to a fixed implementation via DELEGATECALL. A handler flag controls dispatch of reentrant calls.
- Stealth ABBE over BLS12-381: zero-metadata broadcast via blobs (Ethereum Magicians) — A design was proposed that uses EIP-4844 blobs to broadcast private contracts and encrypted state without recipient metadata. It points to the bottleneck that existing broadcast encryption (BE/ABE) requires plaintext headers such as recipient indices, which leaks metadata.
- Sounding out an ERC for AI-evaluated 'Inference Programs' contracts (Ethereum Magicians) — A post was made asking whether the kind of contract discussed in the Solana SIMD repository, which stores prompts and rules and has an AI model evaluate them, could also be standardized as an ERC on Ethereum. It is an extension of the recent flood of agent credit ERC drafts.
- Naming call opens for the J-star name of the consensus layer upgrade after Hegotá (Ethereum Magicians) — A naming call has opened for a name starting with J for the consensus layer upgrade after Hegotá (Heze + Bogotá). The naming convention that ran through Shapella, Dencun, Pectra, Fusaka, Glamsterdam, and Hegotá was also summarized. It pairs with the naming discussion for the execution layer upgrade after Bogotá covered in the previous issue.
Twitter debates
- lex_node and Stani Kulechov criticized Morpho's proposal on the DeFi lending legal framework submitted to the SEC (@lex_node) — lex_node compared the DeFi lending legal frameworks that Aave and Morpho each submitted to the SEC, pointing out that one defends DeFi while the other gives it up through negotiation, and raised the issue that if lending markets or vault deposits are classified as securities purchases, registration and disclosure obligations arise. Stani Kulechov said Aave's reasoning matches the view it used when it successfully defended the past SEC cases involving yearn and Andre Cronje, and agreed that Morpho's submission seems to shift responsibility to curators rather than acknowledging that risk management is the essence of lending. Morpho's rebuttal is not in this collection, so only one side's position has been confirmed.
- Positions split between binji_x, lex_node, and others on whether to call ETH money (@binji_x) — A report (@ETH_MAXER) spread that Vitalik Buterin said at Devcon that ETH will be the most important app on Ethereum and the most valuable asset, and that 'ETH is money' (the primary source is unconfirmed). binji_x agreed that ETH has become a legitimate store of value thanks to the network properties Vitalik has protected, but held back, saying ETH is not yet money and that, in the short term, it is more realistic for overcollateralized ETH stablecoins to create low-volatility dollar-pegged assets that support the role of money. lex_node commented that the ecosystem is aligning toward making ETH a programmable store of value and marketing it that way, and that ETFs, DATs, and investment banks would welcome this.
- Stani Kulechov and lex_node split over the alternative to Paraguay's banking crisis (@lex_node) — Stani Kulechov wrote that more than 60% of the population in Paraguay has lost access to funds deposited in banks, reminded readers of a financial system in which access to one's own money is a basic right rather than an intermediary's privilege, and then wrote 'Bitcoin matters now more than ever.' lex_node rebutted that your project is on Ethereum, and that with ETH one can avoid debanking while borrowing and swapping on that very Aave without intermediaries and even with privacy. The EF's binji_x made a practical case in a long Spanish-language guide, stating that Ethereum cannot recover money already locked in banks, the responsibilities of self-custody and issuers' freezing powers, checking cash-out feasibility and experimenting with small amounts, and Paraguay's crypto asset reporting obligations, and also disclosed his interest in working on Ethereum.
- Kain Warwick doubted the explanation of how Circle froze funds within 3 minutes during the Binance hack (@laurashin) — On Unchained's Uneasy Money episode, Taylor Monahan and Adam Sadowski of LlamaRisk judged that this time Circle also moved to freeze funds 3 minutes after Tether. Kain Warwick doubted the explanation, saying Circle had always claimed it needed presidential-order-level procedure before acting, so how was this possible, and asking 'how stupid do you think we are.' Taylor said Circle claimed it had been contacted by law enforcement, noting that the FBI had been stationed at Binance's offices all weekend, but still admitted it was too fast. In the same episode, Kain pressed Aave on its market sizing, asking whether someone should be able to deposit $1 billion in a single block and borrow all of the ETH. Sadowski answered that Aave is preparing to introduce borrow limits by time period.
- binji_x rebutted gakonst's criticism of the engineering quality at Devcon (@binji_x) — gakonst criticized the seriousness of Devcon's engineering (the original tweet was not collected), and binji_x responded at length. While acknowledging gakonst's contributions, he listed Devcon's engineering programs: a talk on writing an RISC-V assembly EVM with LLMs and formal verification, formal spec verification of Aave V4 spokes using the Certora Prover, Pureth in Reth and a fellowship porting Geth FilterMaps to Reth, a Lean-based standalone EVM spec, zkEVM verification, and a public competition for post-quantum signatures. He partly conceded, however, that there is something to learn from Apple about presentation that makes the work more visible.
Notable remarks
- pcaversaccio: 'Ethereum security does not fund itself' (@pcaversaccio) — pcaversaccio, curator of the TheDAO Security Fund, appealed for funding for the second round of the public security initiative. Pledges are collected only if the initiative is fully funded, and he wrote that if a protocol depends on Ethereum security work, it should put in 10 to 15% and leave the rest of the fundraising to the fund. He cited a formally verified Vyper compiler and a production-grade local-first Safe UI as his own recommendations.
- jchaskin22: 'We promised an Ethereum without gatekeepers, but we planted gatekeepers in almost every path of use' (@jchaskin22) — In a Devcon talk, jchaskin22 pointed out that Ethereum's distinctive properties of censorship resistance, verifiability, and privacy do not reach most users today. The argument is that intermediary layers such as RPCs and frontends have become de facto gatekeepers.
- Barnabé Monnot: In the 3 months since submitting the talk, there was major progress in consensus and post-quantum (@barnabemonnot) — In the Devcon 'Fast Ethereum' session, Barnabé Monnot summarized that in the 3 months after submitting his talk on August 6, the community secured a formally verified next-generation consensus mechanism that enables fast finality and short slots, along with a large improvement in post-quantum cryptography that lowers latency and cost (slides published). The September decoupled consensus finality PoC and the PQTS discussions are thus counted as official results.
- Justin Slaughter: 'Democratic lawmakers who understand crypto think 80% of CLARITY is fine' (@laurashin) — Justin Slaughter of Paradigm said that Democratic lawmakers who understand crypto acknowledge that 80% of the CLARITY bill is fine, and that the DeFi industry must watch very closely as the law about them is being made.
Ecosystem & security
- Liquidation of the Binance hacker's Aave position and preparation of borrow rate limits (@laurashin) — This is a follow-up to the Binance hack ($6B) from the previous issue. The hacker deposited $60 million of USDT into Aave and borrowed ETH. On November 4, on-chain, a liquidator repaid about 9,884 WETH of debt in the hacker's wallet and took about 24.38 million USDT of collateral. Adam Sadowski of LlamaRisk explained that 'nothing was stolen from Aave, the protocol worked as designed, and a market freeze was enforced so that stolen funds would not remain on the books,' and said Aave is preparing to introduce borrow limits by time period. On blame for EIP-7702, security researchers Jonty and pablosabbatella denied it, saying the hack would have happened without 7702 and that it only made things slightly faster.
- The FBI's naming of THORChain node operators and a warning about criminal risk for builders (@laurashin) — Taylor Monahan characterized North Korea's laundering through THORChain and the FBI's naming of node operators by real name as a replay of the Tornado Cash case. Separately from her distrust of THORChain, she said 'I don't want to see builders go to prison.' She added that this is the same kind of warning shot as during the 2022 Ronin hack, and that Roman Storm is still waiting for his next trial. In a separate case, it was also revealed that ZachXBT traced $30 million stolen in a French bank hack, but the funds could not be frozen because the bank did not file a victim report.
- A recovery path for 9 million SNX that an AI proposed in 7 seconds (@laurashin) — Kain Warwick said that while recovering the 9 million stolen SNX, an AI model (referred to as Opus) proposed in about 7 seconds a path the team had not thought of: creating an approval with the token authority permission and transferring the tokens to the DAO. The team had believed the problem could only be solved by changing token state, and at first they objected that it would not work, but they eventually admitted this path was the more elegant solution.
- First permissioned asset listing on the Uniswap AMM, Groma's real estate issuance and redemption (@Uniswap) — GromaCorp put real estate issuance and redemption directly on the Uniswap protocol through the Permissioned Pools Hook. This is the first permissioned asset listing in AMM history and a new standard that enforces compliance checks at the protocol level. Groma's tokenized securities can be used only by accredited investors in the US and by users outside the US.
- Devcon 8 closes, Devcon 9 to be held in Lisbon in October 2027 (@EFDevcon) — Devcon 8 in Mumbai closed on November 6, and it was announced that Devcon 9 will be held in Lisbon from October 11 to 13, 2027. During the event, binji_x warned that an impersonating Telegram account, with the lowercase l in his handle replaced by an uppercase I, had been calling EF researchers, and the privacy track introduced Privacy Pools v2, which lets users do privately everything they can do with a public address. The EF said 2026 was the year the social layer of Ethereum was most meaningfully decentralized in its history, citing new organizations such as ETH Labs as examples of avoiding a single point of failure.
- Stripe Global Payouts goes live on Base (@base) — Stripe's Global Payouts went live on Base, allowing US businesses to send USDC to crypto wallets in more than 160 markets. It is a case of real payment traffic from a major payments provider flowing directly into an Ethereum L2.
Market briefing
- Payward, operator of Kraken, buys AAVE tokens as part of a partnership (@StaniKulechov) — Payward, which operates Kraken, bought AAVE tokens on the secondary market as part of a partnership to expand on-chain lending. The buyers are Payward's affiliated investment fund Digital Asset Special Investments I, LP and its clients, and the possibility of further purchases was also mentioned. Stani Kulechov emphasized single-token concentration under Aavenomics 3.0 and the 'Aave Will Win Framework.'
- Democratic moves to shift crypto jurisdiction to the SEC (@TrustlessState) — After the midterm victory, Maxine Waters's new bill was reported to shift crypto jurisdiction from the CFTC to the SEC, and there is also word that a Democratic version of CLARITY, with stronger SEC oversight and reduced CFTC authority, is being prepared for introduction in the new Congress next year. David Hoffman offered the interpretation that if the SEC holds control, it becomes easier for Democrats to install a 'Gary Gensler 2.0.' Meanwhile, the White House began reviewing a draft SEC rule applying existing broker-dealer rules (15c3-1, 15c3-3, and others) to crypto.
- Kalshi launches perpetual futures on about 50 US stocks (@laurashin) — Kalshi launched perpetual futures on about 50 US stocks. This came two days after CFTC Chair Michael Selig said registered exchanges could be allowed to trade stock perps. The tokenized stock derivatives structure pioneered by on-chain perp protocols is being imported back into a regulated exchange.
Other news
- Uniswap cumulative tokenized stock volume tops $20 billion (@Uniswap) — Uniswap's cumulative Stock Token trading volume exceeded $20 billion, ranking first among all DEXs. Trading has been accumulating quickly since the StablePair Hook was applied to tokenized stock pairs last month.
- Amundi extends tokenization from cash to core European equities (@Etherealize_io) — According to Etherealize, Amundi, Europe's largest asset manager, has extended tokenization from cash to core European equity funds, and the funds' operating model remains unchanged. It was presented as a case of institutional adoption of Ethereum-based on-chain finance.
- SPCXc added to Aave V4 Base market, Base kicks off tokenized stock hackathon (@aave) — SPCXc was added to the Coinbase Tokenized Stocks market on Aave V4, so it can be used as collateral for borrowing USDC in eligible regions outside the US. Base started a month-long tokenized stock product hackathon called 'Stocks Maxxing' with a $100,000 prize pool.
- Arbitrum announces USDG stablecoin growth program (@arbitrum) — Arbitrum announced that it is considering allocating a 165 million ARB incentive program, and a program that pays 200,000 ARB each to the first 10 teams to reach an average USDG TVL of $10 million. GMX's USDG supply grew 10x in a month, from about $1.5 million in early October to about $17 million.
- Optimism: apps on OP Stack chains earned more than $500 million in the first half of 2026 (@Optimism) — Optimism announced that apps running on OP Stack chains earned more than $500 million in the first half of 2026. The figure shows that the profitability of the L2 ecosystem is being demonstrated at the app layer rather than at the chain level.