Anthropic · Claude · Mythos · Ars Technica
Anthropic confirms these topics are too dangerous to let its Fable 5 model talk
Compiled by KHAO Editorial — aggregated from 8 sources. See llms.txt for citation guidance.
✓ KHAO Verified
Anthropic Tuesday publicly released Claude Fable 5, its first “Mythos-class” model that it says surpasses its previous frontier Opus models in overall capabilities.
Key facts
- API and Enterprise users will be able to access the Fable 5 model at a cost of $10-per-million input tokens and $50-per-million output tokens starting today
- Anthropic says Fable 5 operates on the “same underlying model” as Mythos 5, which is coming out of its monthslong “Mythos Preview” period today, but only for “a small group of cyberdefenders” judged
- Anthropic Tuesday publicly released Claude Fable 5, its first “Mythos-class” model that it says surpasses its previous frontier Opus models in overall capabilities
- In over 1,000 hours of red-team testing with a bug bounty program, Anthropic says external teams failed to find any universal jailbreaks for Fable 5
Summary
Anthropic says Fable 5 operates on the “same underlying model” as Mythos 5, which is coming out of its monthslong “Mythos Preview” period today, but only for “a small group of cyberdefenders” judged trustworthy through the existing Project Glasswing. Anthropic said it has tuned these safeguards to be “stricter than ideal,” meaning the system may occasionally refuse “harmless requests” in a way that it acknowledges may be frustrating for regular users. Fable 5’s topic-based safeguards are built around a system of classifiers designed to broadly detect banned prompt subjects as well as any potential jailbreak attempts. The company said it is particularly worried about Mythos 5’s ability to perform “agentic hacking,” executing multi-part cyberattacks with much more facility than earlier models.