In a necessary move to protect national security, Anthropic has been forced to pull its newly released Claude Fable 5 and Mythos 5 AI models from the public.
The company confirmed it had to abruptly disable the tools to ensure compliance following warnings from U.S. authorities regarding potential 'jailbreaking' methods that could allow hackers to bypass security restrictions.
While Anthropic claims the vulnerabilities identified were minor, the reality is that the UK government’s AI Security Institute found the model could exploit defenses and systems 73% of the time—a staggering capability that poses a clear risk to digital infrastructure.
This development comes as the Trump administration continues to hold Big Tech accountable, with Defense Secretary Pete Hegseth previously labeling the company a 'supply chain risk,' a designation typically reserved for adversarial foreign entities.
Anthropic is currently engaged in a legal battle with the Pentagon over this designation, yet the government’s insistence on rigorous security standards remains the right approach in an era where AI tools are becoming increasingly powerful and potentially dangerous.
As the European Union watches from the sidelines, fretting over its own lack of technological sovereignty, the U.S. is prioritizing the safety of its networks over the unchecked expansion of Silicon Valley's latest experiments.
Tags
