TechFlow reported on June 13, citing Cointelegraph, that Zcash founder Zooko Wilcox said a security audit of the Zcash protocol found no new severe vulnerabilities. The review was commissioned by Shielded Labs and carried out using Anthropic’s Mythos AI model, with the audit focused on the Zcash protocol.
The audit followed a previously disclosed issue involving the Orchard shielded pool. Security researcher Taylor Hornby had used Claude Opus 4.8 to identify a four-year-old forgery vulnerability in the Orchard shielded pool. After the issue was found, developers urgently paused Orchard transactions on June 3 and completed the fix on the same day.
According to the report, the Zcash Foundation confirmed that there is no evidence the vulnerability had ever been exploited, and user privacy was not affected. The Anthropic Mythos AI review did not identify any new severe vulnerabilities, adding a further protocol check after the Orchard-related fix.