Anthropic · Claude · Mythos · Bitcoin.com News
Anthropic’s Secret AI Model Mythos Audits Entire Zcash Protocol, Flags No New Bugs
Compiled by KHAO Editorial — aggregated from 4 sources. See llms.txt for citation guidance.
✓ KHAO Verified
Zcash founder Zooko Wilcox-O’Hearn publicly thanked Anthropic this week, confirming the artificial intelligence (AI) company ran a full security audit of the Zcash protocol using its restricted frontier model, Mythos, and found no additional critical vulnerabilities.
Key facts
- As of June 13, ZEC trades in the $411 to $416 range, stabilizing above the $400 support level after recovering 45% to 80% from June 5 lows
- In late May 2026, security researcher Taylor Hornby, working under contract with Shielded Labs, used a custom AI auditing framework built around Claude Opus 4.8 to identify an under-constrained
- The post arrived one week after the public disclosure of a critical soundness bug in the Orchard shielded pool, which had triggered a 38% to 50% price collapse in ZEC within 48 hours of going public
- Anthropic’s Mythos Preview is the company’s most advanced AI model, introduced in April 2026 as part of its Project Glasswing defensive security initiative
Summary
Anthropic’s restricted Mythos AI audited the full Zcash protocol at Shielded Labs’s request and found no new critical bugs as of June 12, 2026. ZEC trades around $411 to $416 on June 13, recovering from June 5 lows near $300 after the Orchard circuit bug was patched via NU6.2. Shielded Labs and Zcash contributors continue security hardening work ahead of the Ironwood upgrade, targeting late July 2026. The statement came directly from Zooko, the online handle for Zooko Wilcox-O’Hearn, computer security specialist, cryptographer, and co-creator of Zcash.