Anthropic, a leading artificial intelligence company, has halted access to its top AI models, Fable 5 and Mythos 5, following a national security directive from the United States government. This directive specifically restricted foreign nationals from accessing these models due to concerns that the systems could be exploited through a technique known as “jailbreaking” for cyber activities. In compliance with the directive, Anthropic decided to withdraw access to the models for all users globally, marking a rare instance where a prominent AI provider has retracted a deployed model due to government intervention.
The decision contrasts with prior US restrictions that have typically targeted semiconductor exports and hardware rather than operational AI systems. Anthropic has expressed disagreement with the government’s decision, arguing that their internal evaluations did not reveal a universal method to bypass the models’ safety features. According to the company, the alleged vulnerabilities do not exceed the capabilities already found in other publicly available AI systems. This incident has sparked renewed debate over the balance between AI innovation, national security, and governmental oversight, emphasizing the strategic importance and control challenges of advanced AI models in global markets.
This development has not gone unnoticed internationally, with significant attention from Europe and India, both of which are major users of advanced AI systems. Industry experts suggest that the situation highlights the necessity for nations to invest in domestic AI research and develop their own sovereign AI capabilities to diminish reliance on foreign-controlled technologies. In India, one of Anthropic’s largest markets, the adoption of AI tools in software, technology services, and enterprise applications is on the rise. Analysts warn that this disruption serves as a reminder of the risks linked with heavy dependence on externally controlled AI infrastructure.
Anthropic has stated that it is actively collaborating with authorities to address the concerns underlying the order and is hopeful of restoring access to the models once these issues are resolved. The company plans to release further technical details about the suspension and the specific security concerns that prompted the government’s action. As this episode unfolds, it is likely to fuel global discussions on AI regulation, technological sovereignty, and the governance of increasingly powerful AI systems.