Crypto Paper Ditches Randomness Beacon in Favor of Distributed Samplers

A new cryptography paper accepted to Crypto 2026 has made headlines for removing a crucial component from its consensus protocol. The 'randomness beacon' was a recurring service that provided fresh public random values at regular intervals, but the new design relies on distributed samplers instead.

The authors of the paper, Damiano Abram, Marshall Ball, Juan Garay, and Aggelos Kiayias, argue that their construction can function without the randomness beacon. They combine proofs of work grounded in fine-grained complexity with multi-verifier signatures of work to produce a secure consensus protocol.

However, the new design still relies on several cryptographic assumptions, including decisional Diffie-Hellman and learning with errors. The authors note that their construction is still theoretical and has not been tested for numerical corruption threshold, synchrony conditions, or communication costs.