0x19958abc1995…19958ab9
Buterin Backs leanPCS as Faster Route to Quantum-Safe Ethereum
Vitalik Buterin now favors leanPCS over leanSTARK for Ethereum's post-quantum upgrade, claiming it delivers the same generality and could ship far sooner. EIP-8288 remains a draft with no mainnet date.

Outputs
Signatures under ECDSA take 65 bytes; quantum-resistant replacements could reach 1-6 KB, with STARK proofs up to 128 KB.
Buterin says a leanPCS PCS-verification feature would cost roughly 20,000-40,000 gas via existing Ethereum execution.
EIP-8288 caps each transaction at 16 signature dependencies and one STARK dependency, with one-second mempool aggregation intervals.
The draft lists a 3,000-gas charge for leanSPHINCS verification and 30,000 for leanSTARK, the latter not yet enabled.
EIP-8288 remains a draft; leanPCS is not yet in the specification and has no hard fork assignment.
Ethereum co-founder Vitalik Buterin has endorsed leanPCS as a potentially faster route to quantum-resistant accounts and efficient on-chain proof verification, in a Monday post on X. The endorsement concerns EIP-8288, the draft improvement proposal he co-authored with Thomas Coratger in June, which would keep large post-quantum signatures and STARK proofs out of Ethereum blocks. Neither the proposal nor leanPCS carries a mainnet activation date; the specification remains incomplete, and Ethereum developers have not approved the approach for any hard fork.
Why does Ethereum need EIP-8288?
The proposal targets a scaling problem that quantum safety creates. Replacing today's elliptic-curve cryptography would swell signatures from 65 bytes under ECDSA to between 1 and 6 kilobytes, while STARK proofs could reach 128 kilobytes. Put plainly, a single quantum-safe signature could take up to roughly ninety times the space of a current one.
"This is unsustainable to put on chain, and 8288 fixes this," Buterin wrote. EIP-8288 would aggregate signatures and STARKs within Ethereum's mempool before transactions reach blocks. Nodes would inspect supporting evidence, produce STARKs confirming specific claims, and combine proofs until a block builder creates one recursive proof covering the relevant transactions. The chain would retain only short claim descriptions; Buterin estimates inter-node evidence at roughly 64 to 96 bytes per claim, plus the aggregate proof. The formal specification describes 96-byte dependency entries and one block-level recursive STARK that validators must check.
The takeaway: Ethereum would verify what was proven, not carry the bulky proof itself.
What changed in Buterin's assessment?
Buterin said his view shifted over recent weeks as he compared three approaches. LeanSPHINCS would aggregate hash-based signatures, suiting quantum-resistant wallets, multisignature accounts and ordinary transfers; a related SPHINCS-minus proposal describes Ethereum-compatible hash signatures verifiable without a new cryptographic primitive, and Buterin suggested users could hold quantum-safe accounts while paying modest gas for each signature's on-chain representation.
LeanSTARK offers wider scope, letting users prove almost any computation, which opens privacy applications, alternative signature schemes and authorization rules that hide sensitive relationships. The obstacle is that developers must first agree on the instruction-set architecture expressing those computations. Buterin called this "the first big case of 'Ethereum moving beyond the EVM'" and warned that settling a hard-to-change programming standard would take time.
LeanPCS sidesteps that decision. A polynomial commitment scheme establishes that a committed mathematical function produces a claimed value, and Buterin said a PCS-verification feature could let developers check complex STARK proofs through existing Ethereum execution at roughly 20,000 to 40,000 gas. "leanPCS gives you the same generality as leanSTARK, in a way that could potentially be implemented very quickly," he wrote, suggesting Ethereum could add leanSTARK later once its programming language is settled. The trade-off: leanPCS may require several on-chain claims per computation and is less compact on-chain, though it grants developers more off-chain flexibility.
In short, Buterin is trading compactness for speed of arrival.
What remains unfinished?
The current EIP-8288 draft has active dependency verification rules only for leanSPHINCS; a leanSTARK verification mode is reserved pending specification. The text lists a 3,000-gas charge for leanSPHINCS and 30,000 for leanSTARK, with the latter not enabled. Transactions are limited to 16 distinct signature dependencies and one STARK dependency, and mempool nodes would aggregate on one-second intervals — restrictions designed to limit abuse of the larger proofs. LeanPCS is not yet written into the specification at all.
EIP-8288 also builds on EIP-8141, a frame-transaction format separating validation, payment and execution, which is scheduled for the Hegotá upgrade but remains a draft without confirmed activation dates. Buterin's endorsement signals direction, not deployment; activation requires a full network upgrade, and no inclusion in a named hard fork has been confirmed.
via crypto.news (Source)
More from Elena Vasquez
Show full bio
Staff writer covering regulation, policy and their impact on crypto and web3 industry at Mempool Brief.
1,525 articles