According to ME AI, on June 13 (UTC+8), Anthropic issued a statement saying that the U.S. government, citing national security concerns, has issued an export control order requiring all foreign nationals to suspend access to Fable 5 and Mythos 5, regardless of their location within or outside the United States, including Anthropic's foreign employees. The practical impact of this order is that the company must immediately disable Fable 5 and Mythos 5 for all customers to ensure compliance. Access to other Anthropic models is unaffected. Anthropic received the government order at 5:21 PM ET today. The letter did not provide details of specific national security concerns. Anthropic interprets this as the government believing it has learned of a method to bypass or "jailbreak" Fable 5. Demonstrations of this specific technique used to identify a small number of previously known, minor vulnerabilities have been reviewed; these vulnerabilities appear relatively simple, and other publicly available models could also be discovered without bypassing them.
Anthropic stated that prior to Fable's release, the company conducted thousands of hours of red-team testing on Fable's security measures with the US government, AISI UK, several private third-party organizations, and its internal team. The tests showed that Fable's security measures are significantly stronger than any previously deployed model, and no testers have yet found a general jailbreak method that can widely bypass the model's security measures. Currently, the government has only provided verbal evidence of a potentially narrow, non-general jailbreak, essentially requiring the model to read specific codebases and fix software vulnerabilities. Anthropic has reviewed the relevant reports and confirmed that the capabilities demonstrated can already be widely provided by other models, including OpenAI's GPT-5.5, and are used daily by defenders to ensure system security.
Anthropic stated that it will comply with government legal directives and remove all user access to Fable 5 and Mythos 5. However, Anthropic disagrees with using a narrow, potential jailbreak discovery as justification for recalling a commercial model deployed to hundreds of millions of people. Anthropic argues that if this standard were applied across the entire industry, it would essentially prevent all leading model providers from deploying new models. Governments should have the ability to prevent insecure deployments, but the process should be transparent, fair, clear, and based on technical facts. Anthropic believes this action failed to adhere to these principles and calls it a misunderstanding; the company is working to restore access as quickly as possible. (Source: ME)
