Scaling quantum-safe signatures
Vitalik Buterin believes leanPCS provides a faster route to quantum security for Ethereum than other proposed methods. He tweeted on Monday that his view on leanPCS shifted from doubt to strong support. Quantum computers threaten to break the elliptic curve cryptography that secures current Ethereum accounts. Replacing these with quantum-safe alternatives creates massive data burdens. Specifically, signatures would grow from 65 bytes to between 1 and 6 kilobytes, while zero-knowledge proofs would balloon from a few hundred bytes to around 128 kilobytes, a scale that is difficult to manage onchain. To solve this, Buterin co-authored EIP-8288 with developer Thomas Coratger in June to aggregate signatures and proofs before they reach a block. Under this design, users publish signatures or proofs, and network nodes repeatedly bundle everything into a single STARK, which is a type of zero-knowledge proof. These aggregate proofs combine until a single "master proof" verifies everything in a block, so nodes only store short claims onchain. In June, Buterin and Ethereum Foundation cryptographers published a post on the Ethereum Research forum regarding SPHINCS-, a family of hash-based signature schemes.
Three paths for EIP-8288
The EIP-8288 proposal provides three ways to handle quantum-safe data depending on the submitted user data through different aggregation methods. These paths vary in complexity and the amount of gas required for verification.
| Path | Functionality | Deployment Speed |
|---|---|---|
| leanSPHINCS | Single hash-based signature type | Fastest |
| leanSTARK | General signatures and privacy protocols | Slowest |
| leanPCS | General signatures and privacy protocols | Fast |
The leanSPHINCS option supports only one type of quantum-safe signature and would cost users a small amount of gas per signature. This path would reduce verification costs from 150,000 gas for current quantum-safe signatures to roughly 3,000 gas. The leanSTARK option allows for any signature scheme, privacy protocols, or hidden multisig logic, but Ethereum must first agree on a new programming language for users’ programs, which Buterin calls "the first big case of Ethereum moving beyond the EVM." This requirement makes leanSTARK slow to implement. leanPCS provides the same generality as leanSTARK but could potentially be implemented very quickly because it requires far fewer design choices, which allows developers to add leanSTARK later. The Ethereum Foundation formed a dedicated post-quantum team in January, and developers have already proposed overhauling the staking deposit contract to support post-quantum validator keys.
