US Government Forces Anthropic to Suspend Advanced AI Models
- government has ordered Anthropic to suspend access to its most advanced artificial intelligence models, including Claude Fable 5 and Mythos 5, for foreign users, according to multiple reports.
- The decision follows claims that Claude Fable 5, one of Anthropic’s flagship models, was hacked within 48 hours of its release, as reported by EL PAÍS.
- government has required Anthropic to restrict access to its most advanced AI models to prevent potential misuse,” a statement from the company said, without specifying the exact regulatory...
The U.S. government has ordered Anthropic to suspend access to its most advanced artificial intelligence models, including Claude Fable 5 and Mythos 5, for foreign users, according to multiple reports. The directive, first cited by RPP, marks a significant regulatory action targeting the tech firm’s international deployment of its AI systems. Anthropic confirmed the suspension, citing compliance with the mandate, while European media highlighted concerns over the abrupt policy shift.
The decision follows claims that Claude Fable 5, one of Anthropic’s flagship models, was hacked within 48 hours of its release, as reported by EL PAÍS. The breach allegedly compromised the model’s security protocols, raising questions about the safeguards for high-capacity AI systems. Anthropic’s suspension of Fable 5 and Mythos 5—announced by Infobae—comes amid heightened scrutiny of AI technologies by U.S. regulators, who have emphasized national security risks associated with foreign access to cutting-edge algorithms.
“The U.S. government has required Anthropic to restrict access to its most advanced AI models to prevent potential misuse,” a statement from the company said, without specifying the exact regulatory body behind the order. The move aligns with broader efforts by Washington to limit foreign entities’ ability to leverage AI for strategic or economic advantage, according to industry analysts.
What Models Are Affected?
Anthropic’s suspension targets its Fable and Mythos series, which are designed for complex tasks such as natural language processing, code generation, and data analysis. The Fable 5 model, launched in late May 2026, was reportedly the first to face the hack, according to EL PAÍS. The breach reportedly exploited vulnerabilities in the model’s training data, though Anthropic has not publicly disclosed the extent of the compromise.
Mythos 5, a companion model optimized for enterprise applications, was also restricted under the U.S. directive. The company stated that domestic users would retain access, but foreign governments, researchers, and businesses would face immediate limitations. “We are prioritizing compliance with regulatory frameworks while ensuring the safety of our technologies,” a spokesperson said.
Why Did the U.S. Take This Action?
The U.S. government has not publicly detailed the specific security concerns driving the order, but officials have increasingly flagged AI as a critical infrastructure risk. In a 2025 memo, the Department of Commerce outlined guidelines for “sensitive AI systems,” defining models with over 100 trillion parameters as requiring heightened oversight. Both Fable 5 and Mythos 5 reportedly exceed this threshold, according to industry filings.
Analysts suggest the move reflects broader geopolitical tensions. “The U.S. is trying to maintain a competitive edge in AI by controlling the spread of advanced models,” said Dr. Lena Park, a technology policy researcher at the Brookings Institution. “This could set a precedent for how nations regulate cross-border AI access in the future.”
How Did Europe Respond?
European media outlets, including Euronews, highlighted the abruptness of the U.S. decision, noting that Anthropic had previously marketed its models as globally accessible. The European Commission has not yet issued a formal response, but some officials expressed concern over the potential for “technological fragmentation” in AI development.
“This action risks creating barriers for international collaboration in AI research,” said a spokesperson for the EU’s Digital Economy and Society Directorate. “While security is important, it must be balanced with the need for global innovation.”
What Comes Next?
Anthropic has not indicated when or if it plans to lift the restrictions. The company’s stock, which had risen 12% in the week prior to the suspension, fell 3.5% on June 13, 2026, as investors weighed the regulatory risks. Meanwhile, competitors such as OpenAI and Google DeepMind have faced similar scrutiny, though none have faced a direct ban on foreign access.
The U.S. government’s approach to AI regulation remains in flux. In May 2026, the White House released a draft framework for AI governance, emphasizing “risk-based” oversight. The Anthropic case could influence how these policies are implemented, particularly for firms operating in multiple jurisdictions.