Anthropic, an AI company, has been unsuccessfully lobbying the US government to lift a ban on its new AI program, Fable. The ban, imposed last week, restricts access to Fable globally, preventing its use by non-Americans. US officials expressed concerns that countries like Russia and China could exploit Fable’s capabilities – designed to identify software vulnerabilities – for malicious purposes. Anthropic maintains Fable is intended to help security professionals fix weaknesses in computer programs and has built-in safeguards against misuse. However, a non-public Amazon investigation reportedly revealed these safeguards could be easily circumvented, raising alarm within the White House. Despite ongoing discussions, including in-person meetings in Washington, the US government remains firm in its decision, believing the risk of misuse is too high. Anthropic continues to assert the safety of its AI and its potential benefits.