Vitalik Buterin Calls for Research on Poseidon Hash for Ethereum’s Future

Highlights:

• Ethereum is testing the Poseidon hash function to improve zero-knowledge proofs.
• The Pectra upgrade failed to go live on the Holesky testnet due to a contract address error.
• Developers plan to test Pectra on Sepolia next, but delays may happen if more bug fixes are needed.

Ethereum is exploring the Poseidon hash function to improve zero-knowledge-proof capabilities. Vitalik Buterin has asked researchers to analyze Poseidon’s security. In support of this, Ethereum has extended the funding application deadline to March 15 to collect more information before deciding on its implementation.

Highly encourage researchers to participate in the Poseidon cryptanalysis program.

We are seriously considering migrating Ethereum to the Poseidon hash to optimize zk-prover friendliness, so having more information about its security properties is extremely high value. https://t.co/AhbV8abD4z

— vitalik.eth (@VitalikButerin) February 26, 2025

Ethereum introduced the Ethereum Improvement Proposal (EIP) 5988 in November 2022. It proposed to add a precompiled contract to the Ethereum Virtual Machine. The goal was to incorporate Poseidon into the EVM hash function to enable zero-knowledge rollups to respond better to the EVM.

Developers introduced Poseidon in 2019, and it is newer than traditional hash functions like SHA-256 and Keccak. These older functions have been subjected to decades of use and testing. However, they were not specifically designed for zero-knowledge proofs. Poseidon was created for this purpose, which qualifies it for Ethereum’s future upgrades. Although researchers have not tested it as extensively as SHA-256 and Keccak, Ethereum’s layer-2 systems already use it.

Pectra Upgrade Faces Setback on Holesky Testnet

Apart from Poseidon, Ethereum is also working on different improvements. The Pectra upgrade is slated to go live on April 8 and is supposed to help to improve the ease of use of the network and layer 2 efficiency. This upgrade follows the Dencun hard fork, which reduced transaction fees in 2024. Pectra will also introduce account abstraction tools, which could make wallets simpler and safer.

Developers tested Pectra on the Holesky testnet before its mainnet launch. Despite going live, the upgrade was not finalized. Security expert Raul Riesco discovered an error in indicating the precise, smart contract address for Pectra Request Hash Calculations. This mistake caused the upgrade to not lock in and create disruption.

[Ethereum Pectra upgrade not finalizing]

It seems that Holesky upgrade failure was due to a simple bug, the team forgot to specify the proper smartcontract address which is used for Pectra Request Hash Calculations.

Fixhttps://t.co/pRzIxfZEEu https://t.co/oJ2dgBOa8w

— Raul Riesco, PhD (@rriescog) February 25, 2025

Holesky is a testing network, so the issue will not affect Ethereum users because the main blockchain is still working. The failure, however, has also called into question whether the Pectra upgrade will be ready for deployment.

Ethereum had faced challenges in the previous testnets before they were implemented successfully. In particular, the Merge network upgrade experienced test execution failures before it launched on the main network. Holesky testnet replaced older networks to better simulate a real environment. Ethereum developers use Holesky to detect any problems before upgrades reach the mainnet.

What’s Next for Ethereum’s Upcoming Upgrade?

Ethereum developers have scheduled the next testing phase for Pectra on the Sepolia testnet. Initially, developers planned to launch it on March 5, but Holesky’s failure may cause delays. If developers need additional time for bug fixes, the Sepolia test could be postponed.

The mainnet activation in April might depend on how smoothly the two testnets run. This is not the first time that Ethereum has run into testnet problems, as the same happened with the Dencun upgrade in January 2024. The upgrade failed to finalize on the Goerli testnet before succeeding later on.