Research Identifies Vulnerabilities in Fiat-Shamir Transform
/ 1 min read
🔍 New research reveals vulnerabilities in the Fiat-Shamir transform for interactive proofs. A study by Ron D. Rothblum and Lev Soukhanov highlights significant flaws in the widely used Fiat-Shamir (FS) transform, which converts interactive protocols into non-interactive ones. While the FS transform is sound in the random oracle model, the authors demonstrate that it can fail when applied to a popular interactive succinct argument based on the GKR protocol. They provide a method to generate valid proofs for false statements, indicating that security guarantees depend heavily on specific circuit implementations. The research also explores variations of the attack that compromise non-adaptive soundness, although these require larger circuits or additional assumptions about cryptographic primitives.
