United States: Anthropic has suspended its newly released AI models, Claude Fable 5 and Mythos 5, after US authorities raised national security concerns just days after their public rollout.
In a statement published on its website, the company said it was instructed to block foreign nationals from accessing Claude Fable 5. As a result, Anthropic said it had no choice but to abruptly disable Fable 5 and Mythos 5 for all customers to comply with the directive.
Jailbreak concerns trigger intervention
According to Anthropic, US national security officials did not cite specific threats but pointed to a potential method for bypassing the model’s safety restrictions, commonly referred to as jailbreaking.
The company said it reviewed a demonstration of the technique and found it had only exposed a small number of previously known, minor vulnerabilities. It also added that similar weaknesses could reportedly be identified using other publicly available AI models.
The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees.
The net effect of…
— Anthropic (@AnthropicAI) June 13, 2026
Jailbreaking refers to attempts to bypass built-in safeguards in software systems, potentially allowing access to restricted functions or sensitive outputs.
Model described as too powerful
Claude Fable 5 is part of Anthropic’s Claude family of AI systems, positioned to compete with OpenAI’s ChatGPT and Google’s Gemini. The company had previously described the model as too powerful, noting that its capabilities exceeded all prior public releases.
Ahead of its launch, Anthropic had provided limited early access to select organisations to test security risks, citing concerns that the model could potentially be misused for cyber exploitation.
Political and legal tensions escalate
The suspension comes with growing friction between Anthropic and the US government. The company is currently involved in legal proceedings with the Trump administration over restrictions affecting the use of its AI systems by government agencies.
Separately, US Defence Secretary Pete Hegseth previously labelled Anthropic a ‘supply chain risk,’ a designation typically reserved for companies from geopolitical adversaries. A federal judge has since ruled that the directive cannot be enforced while litigation continues.
Anthropic has also challenged the Pentagon’s designation in court, arguing against restrictions placed on its technology.
Industry scrutiny intensifies
The incident highlights increasing scrutiny of advanced AI systems as governments weigh innovation against national security risks. While Anthropic maintains that no new critical vulnerabilities were discovered, regulators appear to be taking a precautionary stance toward increasingly capable models.
Further updates from US authorities and Anthropic are expected as the legal and regulatory situation develops.
MOST READ | Visa denial forces Thomas Partey to miss Ghana’s World Cup opener
