WiretechnologyBrief
Aztec’s Zac Williamson ‘Can’t Rule Out’ Newer AI Models Finding Bugs in Aztec Tierra
Zac Williamson said Aztec has formally verified only its most security-critical code, and that, for now, every major upgrade makes users withdraw and redeposit so a hidden double-spend can't carry over.

Brief
Zac Williamson, co-founder and CEO of the Aztec Foundation, told Unchained in an interview recorded Nov. 9 that more capable AI models may still uncover bugs in Aztec Tierra (Alpha V6), the privacy network's newest release, which launched Nov. 3. Automated audits of the previous version reportedly turned up several severe issues in its cryptographic libraries. Aztec earlier said AI-assisted reviews of V5 found one critical proving-system flaw that could pass a forged proof, and it could not determine whether anyone exploited it. Williamson said the team has formally verified the subsystems where the AI found bugs, but the protocol as a whole is not yet fully verified. He also said major upgrades run as full hard forks, so users must withdraw and redeposit funds, and that the network has no security council. ETHBEAT rates V6 Stage 2 and lists it as starting without state carried over from V5.
Why it matters: It shows how a privacy network's team weighs AI-driven attack risks against formal verification and upgrade design, which bears on user funds.
Read the full story
Dextape context
Built from Dextape’s own wire, stories and glossary.
Dextape story on this event
Mentioned in this story
Open a coin page for its current price.