#Ethereum Tech Upgrade
Ethereum Pivots From Poseidon To SHA/BLAKE As Proof Speeds Surge 9x
WooFun2026-08-15 07:40
Key Takeaways
Justin Drake confirms Ethereum Foundation is shifting from Poseidon to SHA/BLAKE for future L1 designs. This leverages faster proof systems for leanXMSS signatures and leanVM aggregation, requiring no immediate user action or rollup changes.
Woofun AI reports that Justin Drake announced on Aug. 13 the Ethereum Foundation's strategic pivot away from Poseidon toward SHA or BLAKE hashes for future layer-1 designs, ending an eight-year commitment to specialized proof-friendly cryptography.
The deeper driver is a fundamental shift in performance tradeoffs, as recent gains in proof systems have rendered specialized hashes less critical for speed. Flock's authors demonstrate that one M4 Max core can prove approximately 82,000 BLAKE3 compression evaluations, 42,000 SHA-256 compressions, and 30,000 Keccak permutations per second. With ten cores exceeding 660,000 BLAKE3 compressions per second, SHA-256 proving ran more than nine times faster than Binius64 in the paper's comparison.
Woofun AI data shows this efficiency allows Ethereum to prioritize the extensive cryptanalytic history of conventional hashes over the raw proving speed of younger primitives.
Structurally, this transition supports a roadmap to replace validator BLS signatures with hash-based leanXMSS signatures, subsequently using leanVM to aggregate their larger data into compact proofs. Cheap proofs of conventional hashes remove a significant obstacle from this architecture, enabling a more robust future proof-driven L1 without compromising security assumptions derived from years of public analysis.
Notably, this decision involves no break in Poseidon, no migration order, and no changes to existing rollups or zkVMs. With no deployed fork and wallets remaining unchanged, the shift concerns only which hash foundations may be utilized in future designs, marking a calculated evolution in cryptographic strategy rather than an immediate operational disruption.
Comments
No comments yet.