Anthropic Suspends Claude Fable 5 and Mythos 5 AI Models Following US Government Security Order

Anthropic has taken the unprecedented step of suspending its newly released AI models, Claude Fable 5 and Mythos 5, following urgent security concerns raised by U.S. authorities. The decision, which comes just days after the public launch of the tools, was prompted by a government order requiring the company to prevent foreign nationals from accessing the technology. Described by officials as "too powerful," the models were abruptly disabled to ensure full compliance with national security protocols and to mitigate potential risks associated with the high-level capabilities of these systems.

While the U.S. government has not publicly identified specific technical vulnerabilities, officials have pointed to potential "jailbreaking" threats that could allow unauthorized users to bypass existing safeguards and access sensitive features. Anthropic maintains that it had already implemented various security measures and conducted limited release testing to identify weaknesses before the broad launch. However, the perceived risk of exploitation by foreign entities led to the immediate disablement of the systems. The company acknowledged awareness of potential vulnerabilities that could be exploited but emphasized its commitment to regulatory compliance over immediate commercial availability.

The suspension is set against a complex legal backdrop, as Anthropic is currently embroiled in litigation with the Trump administration regarding its designation as a security risk. Despite this friction, a federal judge recently ruled that government agencies may continue using Anthropic’s tools while the legal proceedings remain ongoing. Meanwhile, industry experts have expressed concern that such heavy-handed restrictions might backfire. Critics argue that limiting access and halting usage could stifle international collaboration and hinder the development of robust cybersecurity frameworks essential for the safe evolution of artificial intelligence.

As the situation develops, the suspension of Claude Fable 5 and Mythos 5 highlights the growing tension between rapid AI innovation and national security imperatives. The outcome of Anthropic's legal challenges and its ability to satisfy government security requirements will likely set a significant precedent for how high-capability AI models are regulated in the United States. For now, the global AI community remains watchful of how these restrictions will impact future research and the delicate balance between technological openness and state-level protectionism.