Anthropic has announced that it was forced to suspend access to its advanced Fable 5 and Mythos 5 AI models after receiving an emergency US government export control order citing national security concerns. The directive reportedly bans access for all foreign nationals, prompting Anthropic to disable the models for all users to ensure compliance.
The decision appears linked to a discovered jailbreak technique that could bypass safety measures and help identify software vulnerabilities. Anthropic strongly disputes the move, arguing the issue is limited, similar capabilities exist in other AI models, and complete jailbreak resistance is currently impossible.
The company warned that such government intervention could hinder innovation across the AI industry and create significant business risks for US tech firms. Anthropic says it is working with officials to resolve the situation and restore access to the affected models.
Source: Android Authority
