Ethereum is shifting away from the Poseidon hash function after new research made traditional hashes like SHA-256 and BLAKE more practical for future network upgrades, potentially impacting protocol design and post-quantum security planning.
Ethereum is preparing to move away from the Poseidon hash function in its future layer-1 protocol designs, according to recent statements from Ethereum Foundation researcher Justin Drake. The decision follows eight years of work on specialized cryptography intended to make zero-knowledge proofs more efficient, but new advances in proof systems have changed the calculus for which hash functions are most practical for the network's long-term roadmap.
Proof System Advances
Zero-knowledge proofs, such as SNARKs, allow for compact verification that a computation was performed correctly. Historically, these proofs have favored arithmetic over large prime fields, making conventional hash functions like SHA-256 and Keccak expensive to prove. Poseidon was introduced to address this by aligning better with the math used in these proof systems, reducing computational costs for Ethereum rollups and privacy applications.
However, recent research into binary-field proof systems has shifted the performance landscape. According to EgonCoin, new techniques now allow for much faster proofs of traditional hash functions. For example, the Flock research paper demonstrated that a single M4 Max core can process over 80,000 BLAKE3 hash compressions per second and more than 40,000 SHA-256 compressions per second. These improvements mean that the proving cost penalty for using established hash functions is no longer as significant as it once was.
Security and Design Implications
Drake emphasized that the decision to move away from Poseidon is not due to any discovered vulnerability or break in the hash function. Instead, the shift is driven by the improved efficiency of proof systems that now make conventional hashes viable for Ethereum's future architecture. This change allows Ethereum to consider hash functions with a longer history of public analysis and cryptanalysis, such as SHA-256, which is part of the NIST Secure Hash Standard, and BLAKE2, which has been publicly specified since 2015.
By adopting more widely studied hash functions, Ethereum can potentially reduce its attack surface and simplify protocol security. The network's security roadmap already identifies protocol simplification as a key strategy for minimizing risk. Faster proofs of conventional hashes also open the door for Ethereum to integrate post-quantum signature schemes and other cryptographic primitives that were previously impractical due to performance constraints.
Roadmap and User Impact
Ethereum's long-term roadmap includes plans to replace validator BLS signatures with hash-based leanXMSS signatures, which are designed to be resistant to quantum attacks. The network aims to use a new virtual machine, leanVM, to aggregate these larger signatures into compact proofs. The ability to efficiently prove conventional hashes removes a major technical barrier to this architecture.
Drake's personal timeline suggests that a production-ready leanVM could arrive around 2027, with deployments across Ethereum's consensus, data, and execution layers by 2028. The official post-quantum roadmap from the Ethereum Foundation targets 2029 for core infrastructure upgrades, with full migration of the execution layer expected to take longer. For now, users are advised that no immediate changes are required to their wallets or existing rollups, as the decision only affects future protocol design rather than current deployments.
According to the Flock benchmark results published in July 2025, binary-field proof systems can now process SHA-256 proofs more than nine times faster than previous approaches, with BLAKE3 and Keccak also seeing significant throughput gains. These findings suggest that Ethereum's pivot to conventional hashes could support higher transaction throughput and more robust security as the network evolves.
Ethereum's shift away from Poseidon reflects a broader trend in blockchain protocol design, where advances in cryptographic research can rapidly alter the trade-offs between performance, security, and future-proofing. As the network prepares for post-quantum threats and seeks to streamline its infrastructure, the choice of hash function will remain a critical factor in both technical and security planning.
Hash functions are a foundational component of blockchain security, serving as the backbone for transaction integrity, consensus mechanisms, and cryptographic proofs. The choice of hash function affects not only performance and scalability but also the network's resilience to future cryptographic attacks, including those from quantum computers. As Ethereum and other blockchains adapt to new research and evolving threats, the ability to update core cryptographic primitives without disrupting users or compromising security will be essential for long-term sustainability.
