TLDR Tech / Dev
More details on Fable 5's cyber safeguards and our jailbreak framework
Fable 5 is being redeployed with stricter cybersecurity classifiers that block prohibited and high-risk dual-use cyber requests while allowing clearly defensive uses. Anthropic uses a Cyber Jailbreak Severity scale to rate jailbreaks by capability gain, breadth, weaponization, and discoverability.
Read the full story