Aztec Discloses a Critical Soundness Flaw in Its Alpha V5 Proving System

Aztec, the privacy-focused Ethereum layer-2 network, has disclosed a critical vulnerability in the proving system behind its Alpha V5 release. According to the project's disclosure, an attacker could construct a proof that passes verification for a transaction the network is supposed to reject, and an accepted proof of that kind would move the chain into a state its own rules forbid. Contributors say they found the flaw through internal auditing assisted by artificial intelligence, that they cannot determine whether anyone exploited it beforehand, and that users should treat funds, applications and contract state on V5 as exposed until operators complete the required network actions. The fix is scheduled for the V6 release, including circuit changes that stop the network accepting proofs tied to the affected system.

This is a different failure class from a smart-contract exploit. On an optimistic rollup, honest validators re-execute transactions and can catch an invalid state transition. A zero-knowledge system replaces re-execution with a mathematical argument, so if the argument itself is unsound, there is no additional check inside the protocol to catch the error. On a privacy network the gap is wider, because the data needed to independently check balances is encrypted by design. Soundness is not one control among several. It is the control.

The disclosure pattern is now familiar. Aztec published a comparable critical vulnerability notice for Alpha v4 and has raised its bug bounty to $2 million. Two critical proving bugs surfacing in consecutive alpha releases are evidence that the review process is working as intended on pre-mainnet software, and also that proving stacks are shipping faster than anyone can fully verify them.