Vitalik Buterin turns bullish on leanPCS in Ethereum’s EIP-8288 push
Ethereum co-founder Vitalik Buterin has endorsed leanPCS as a possible faster route to quantum-resistant accounts and efficient cryptographic proofs, explaining how EIP-8288 could keep large verification data out of Ethereum blocks.
- Buterin has backed leanPCS as a potentially faster approach to Ethereum’s quantum-safe cryptographic infrastructure.
- EIP-8288 proposes aggregating large signatures and STARK proofs before transactions enter Ethereum blocks.
- Buterin believes leanPCS could support flexible cryptographic applications without immediately introducing a new programming architecture.
- LeanSPHINCS targets quantum-safe account signatures, while leanSTARK would support more complex privacy and verification applications.
- EIP-8288 remains a draft, and the leanPCS approach has not been approved or assigned a mainnet activation date.
As previously mentioned, Buterin noted in a Monday X post that his view of leanPCS had changed over recent weeks. He compared three approaches for handling signatures and proofs, focusing on a problem Ethereum would encounter when replacing current cryptography with quantum-resistant alternatives: the new signatures and proofs take up considerably more space.
“This is unsustainable to put on chain, and 8288 fixes this,” he wrote, referring to signatures that could grow from 65 bytes under ECDSA to between 1 and 6 kilobytes, and STARK proofs that could reach 128 kilobytes.
EIP-8288 Would Combine Cryptographic Proofs Before They Reach Blocks
The EIP-8288 draft, written by Buterin and Thomas Coratger in June, proposes aggregating signatures and STARKs within Ethereum’s mempool, where pending transactions circulate before entering blocks. Instead of sending each large proof through the chain, users would publish the evidence needed to support a transaction or computation.
Nodes would examine the evidence and produce a STARK confirming specific claims, such as whether a public key signed a message or a program returned a particular result. Subsequent nodes could combine those proofs again until a block builder creates one recursive proof covering the relevant transactions.
Ethereum would retain short descriptions of the verified claims on-chain, with the full underlying signatures and proofs handled outside ordinary block calldata. Buterin estimated the information being passed between nodes at roughly 64 to 96 bytes per claim, alongside the aggregate proof.
As previously reported, the earlier EIP-8288 proposal sought to move heavy cryptographic verification away from repeated execution inside the Ethereum Virtual Machine. The formal specification describes 96-byte dependency entries and one block-level recursive STARK, which validators would need to check.
Buterin Now Sees LeanPCS as a Faster Route Than LeanSTARK
The three approaches Buterin compared differ in what users submit for verification. LeanSPHINCS would aggregate a particular kind of hash-based signature, making it suitable for quantum-resistant wallets, multisignature accounts and regular transfers.
A separate SPHINCS-minus research proposal describes Ethereum-compatible, hash-based signatures that can be verified without introducing an entirely new cryptographic primitive. Buterin said leanSPHINCS could eventually let users make quantum-safe accounts while paying only a small amount of gas for each signature’s on-chain representation.
LeanSTARK would offer more freedom. A user could prove that almost any computation was performed correctly, opening the design to privacy applications, different signature schemes and account authorization rules that hide sensitive relationships.
The difficulty is that Ethereum developers would first have to settle on the instruction-set architecture (ISA) used to express those computations. Buterin called this “the first big case of ‘Ethereum moving beyond the EVM’.”