The Ethereum Foundation is moving away from Poseidon as a planned hash function for a future Ethereum base-layer design, Ethereum Foundation core researcher Justin Drake said on Aug. 13, 2026.

Drake said advances in proof systems that operate over binary fields have changed the trade-off between conventional hash functions and specialized SNARK-friendly hashes such as Poseidon.

Why Ethereum Is Moving Away From Poseidon

Poseidon is a hash function designed specifically to be efficient inside SNARK proofs, and it became Ethereum’s dominant SNARK-friendly hash after its 2019 release.

According to Drake, newer binary-field proof systems can handle the operations used by conventional hashes more efficiently, which changes the engineering case for keeping a specialized hash in the future base layer.

In a post on X, Drake wrote: “Goodbye, Poseidon! An epic 8-year, 8-figure rabbit hole in post-quantum cryptography reaches its dream conclusion. The Ethereum Foundation is abandoning Poseidon for L1, pivoting to SHA or BLAKE. This milestone unlocks ultimate security for lean Ethereum and foreshadows a golden…”

He credited Binius research by Benjamin Diamond and Jim Posen, along with newer work called Flock, with helping change the trade-off.

Drake said these systems can prove roughly 1 million conventional hash computations per second on a laptop, about 100 times slower than native CPU computation.

That performance, he said, could make established hashes such as SHA-2 and BLAKE more practical for Ethereum’s future base-layer design, reducing the need for a specialized SNARK-friendly hash.

The Shift Fits Ethereum’s Post-Quantum Strategy

Drake also linked the proposed direction to Ethereum’s broader post-quantum strategy.

He argued that Ethereum should favor established cryptographic assumptions and remain cautious about newer cryptographic constructions. Drake cited setbacks involving HAWK, a lattice-based signature scheme, and SQIsign, an isogeny-based signature scheme, as examples of the risks associated with newer cryptographic approaches.

The broader strategy includes preparing Ethereum’s cryptographic infrastructure for the possibility of future quantum computers, which is why the discussion centers on long-term base-layer design rather than a short-term application change.

Poseidon Change Targets Ethereum’s Future Base Layer

The proposed change does not make Poseidon obsolete for existing applications and does not require developers to migrate immediately.

Instead, it concerns Ethereum’s future base-layer architecture.

Drake said a production-grade leanVM is targeted for 2027, with consensus, data, and execution-layer deployments targeted for 2028.

The timing makes the hash-function decision part of a longer rollout path for Ethereum’s next-generation infrastructure, rather than an immediate protocol update.