Artificial intelligence company Anthropic has suspended access to its newly released Claude Fable 5 model after US authorities raised security concerns just days after its public launch.
The company announced the decision in a statement, revealing that it had been instructed to restrict access to the powerful AI system for foreign nationals. To comply with the order, Anthropic temporarily disabled both Claude Fable 5 and Claude Mythos 5 for all users.
Why Was Claude Fable 5 Suspended?
According to Anthropic, government officials informed the company about a potential method of bypassing, or “jailbreaking,” the model’s safeguards. Jailbreaking refers to techniques that allow users to circumvent built-in restrictions, potentially exposing vulnerabilities or enabling unauthorized actions.
However, Anthropic stated that the vulnerabilities identified were relatively minor and already known within the cybersecurity community. The company also noted that similar issues could reportedly be discovered using other publicly available AI models.
A Powerful AI Model Under Scrutiny
Claude Fable 5 is one of Anthropic’s most advanced AI systems and was designed to compete with leading models such as ChatGPT and Gemini. Before its public release, Anthropic described the model as exceptionally capable and implemented additional safeguards to prevent misuse.
The company initially provided limited access to selected organizations for testing and security evaluations, citing concerns that the model’s advanced capabilities could potentially be exploited for cyberattacks or unauthorized system access.
Legal Battle With the US Government
The suspension comes amid an ongoing legal dispute between Anthropic and the US government. The company is currently challenging restrictions that sought to limit the use of its AI products by government agencies.
The controversy intensified after US officials reportedly classified Anthropic as a “supply chain risk,” a designation rarely applied to American technology companies. Anthropic argues that the classification is unjustified and has challenged the decision in court.
A federal judge has temporarily blocked enforcement of the directive, allowing government agencies and defense contractors to continue using Anthropic’s AI tools while the case proceeds.
Umar Daraz is the Founder and Editor of TodayEast, an independent digital news platform covering global affairs, business, technology, sports, and world rankings. He specializes in data-driven journalism, international news analysis, economic trends, and emerging technologies. Through TodayEast, he aims to provide accurate, accessible, and insightful reporting on the stories shaping the world.