BlockBeats News, August 13th. Ethereum Foundation core researcher Justin Drake announced that Ethereum L1 will abandon the SNARK-friendly hash function Poseidon, which has been dominant since 2019, and instead adopt traditional hash functions such as SHA2 or BLAKE2s. This reversal stems from a breakthrough in SNARK design—the key lies in being "hash-friendly to SNARK" rather than "SNARK-friendly to hash." By natively aligning binary field operations with traditional hashes, the proof performance of traditional hashes in SNARK has reached 1 million proofs per second with only about 100x overhead, in stark contrast to the previously extremely expensive bitwise operations in large prime fields. Drake referred to this as "science fiction-grade cryptography" and specifically mentioned contributors to breakthrough research such as Binius and Flock.

This transition will bring Ethereum's minimal assumptions in hash-based cryptography to the peak. Deployment speed will be greatly accelerated, eliminating the need to wait for years for Poseidon's cryptographic analysis to mature. The EF Post-Quantum team is rapidly advancing, with a roadmap leading to the deployment of leanVM in 2027, and the consensus, execution, and data layers in 2028.

Drake also pointed out that AI's enhanced capabilities in cryptographic analysis have recently dealt blows to lattice-based schemes (HAWK) and isogeny-based schemes (SQIsign), while hash-based schemes are emerging as the victors in post-quantum blockchain signatures. The trend of open-source automated research is also accelerating, with SNARK.fast achieving 1.8 million BLAKE3 proofs per second.