Anthropic ships Claude Fable 5.2 and Mythos 5.2
Original title: Anthropic’s Claude Fable 5.2 blocks fewer security tasks, while Mythos gets better at finding bugs
Fable 5.2 cuts unnecessary cybersecurity blocks by about 60%, while Mythos 5.2 finds more bugs than Opus 5.5.
Anthropic released Claude Mythos 5.2 to selected partners and Fable 5.2 to regular users, citing coding and reasoning improvements. It calls Mythos its most capable cybersecurity model; on Anthropic’s CVE Rediscovery evaluation it detects 10.2% more vulnerabilities than Claude Opus 5.5 and 14.4% more than Mythos 5.1. Fable still blocks penetration testing, exploit generation, and some binary analysis, switching those replies to Opus 4.8. Anthropic says unnecessary cybersecurity blocks fell about 60% versus Fable 5.1.
Unconfirmed: The vulnerability-finding gains and the one-in-six claim are Anthropic’s figures or a partner report to Anthropic, not independently verified.
- A partner told Anthropic one in six confirmed Mythos bugs were missed by other frontier models.
- High-risk Fable cybersecurity activity is automatically switched to Opus 4.8.
- The 60% figure is from early testing on Claude.ai, Claude Code, and Cowork.
- Anthropic says its evaluations cannot catch every failure before release.
- Organizations
- Anthropic
From the original article
2 min read at bleepingcomputer.com
Anthropic recently rolled out Claude Mythos 5. 2 to selected partners and Fable 5. 2 to regular users, with improvements to coding and reasoning.
By Mayank Parmar · bleepingcomputer.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.