Anthropic has released Claude Fable 5, its most advanced AI model to date, featuring integrated cybersecurity safeguards that route risky requests to a less capable model to prevent misuse. Alongside Fable 5, Anthropic offers Claude Mythos 5— the same powerful model without these restrictions—exclusively for vetted cybersecurity professionals to safely leverage its exploit-finding capabilities. This dual-product approach addresses the risk of malicious use while supporting defenders in vulnerability discovery and patching, highlighting the evolving challenges and strategies in securing AI-driven software vulnerability management.
https://thehackernews.com/2026/06/anthropic-releases-claude-fable-5-its.html