Support for Ethereum’s Shift
Adam Back, the CEO of Blockstream and the mind behind Hashcash, has expressed his support for Ethereum’s recent decision to pivot towards more traditional cryptographic standards. His remarks come in light of Ethereum’s halt on the Poseidon hash function project, a significant initiative that spanned nearly eight years and substantial investment. Instead, Ethereum will revert to established cryptographic methods like SHA and BLAKE2s for its Layer 1 protocols.
Concerns Over Custom Algorithms
This decision was announced by Ethereum Foundation researcher Justin Drake, prompting Back to share his skepticism regarding the reliance on custom, ZK-optimized algorithms that, in his view, lack rigorous scrutiny from the broader technical community. Back pointedly remarked,
“Never liked prover-friendly hashes anyway,”
underscoring his long-standing belief in the importance of using well-vetted cryptographic techniques. For him, the pursuit of maximum reliability has always taken precedence, declaring a preference for the higher computational expense associated with standard hash algorithms rather than the risks posed by less mature technologies.
Technological Advancements
Historically, the computational intensity of conventional cryptography posed challenges for zero-knowledge systems, prompting a rush to develop bespoke solutions like Poseidon. However, Back’s cautious stance has gained momentum thanks to new advancements in binary-field mathematics. The introduction of the Binius proof system in 2023 and the anticipated Flook innovation in 2024 significantly enhance the processing capabilities, allowing up to a million standard hashes to be calculated per second on ordinary devices.
Future Plans for Ethereum
This evolution in technology has effectively undermined the justification for Poseidon, leading Ethereum to reconsider a substantial amount of development time and resources previously allocated to it. Looking ahead, the Ethereum Foundation has laid out plans for incorporating advanced post-quantum security measures, starting with the rollout of the LeanVM virtual machine in 2027, aiming for full integration at the network’s base level by 2028.
