Claude Fable 5 Returns After US Ban: New Safety Layer Added

Claude Fable 5, Anthropic's flagship AI model, is back online after a US ban. The model was removed just three days after its introduction due to government concerns about its use in cybersecurity. What's changed is a new protective layer. Anthropic developed a classifier to detect prompts aimed at uncovering software vulnerabilities or generating exploit code. If detected, these requests are redirected to an older model, Claude Opus 4.8. This safeguard reportedly blocks bypass techniques in over 99% of tests. The underlying model itself has not fundamentally changed. The company says the model is now available through Claude.ai and other platforms, with cloud access being restored separately. This incident highlights ongoing debates within the AI industry regarding regulation and safety.