The Trump administration has lifted restrictions on Anthropic’s latest Claude chatbot models, ending a weekslong ban triggered by cybersecurity concerns.
Anthropic announced Tuesday night that Claude Fable 5 is now widely available. Its most powerful model, Mythos 5, is also being restored — but only for a select group of US-based organizations approved by the federal government.
The Commerce Department blocked foreign nationals from accessing both models on June 12. Anthropic had to pull the products down for all users just days after launching them. The government’s concerns were sparked by a report from cybersecurity researchers at Amazon, Anthropic’s cloud provider. They found a way to bypass Fable 5’s safety guardrails, potentially letting it discover and exploit software vulnerabilities.
This isn’t out of nowhere. Earlier this year, Anthropic itself warned that Mythos was good at finding software flaws in ways that could be weaponized by malicious hackers. Officials have been increasingly nervous about AI models that can automate vulnerability discovery at scale.
OpenAI is facing similar pressure. The company said Friday it’s restricting its new GPT-5.6 Sol model at the administration’s request, limiting access to approved customers for a temporary period.
Last month, Trump signed an executive order on AI oversight that set up a framework for the federal government to vet national security risks of advanced AI systems for up to 30 days before public release. Participation is technically voluntary, but the message is clear: AI safety is now a government affair.
