The United States government has rescinded the export ban imposed on Anthropic's most advanced artificial intelligence (AI) tools, namely Claude Fable 5 and Mythos 5, following assurances from the tech company that it will proactively address security risks associated with these models. Initially halted in June due to potential national security concerns, the Commerce Department has now confirmed that Anthropic has addressed the risks, as stated in a letter obtained by the BBC. The advanced AI tools, which were designed for both consumer and business markets, were released on June 9th. Despite initial concerns about potential vulnerabilities, Anthropic had previously maintained that US authorities had not identified specific issues with their technology. The Commerce Department has reserved the right to reassess its decision if necessary in the future. Anthropic's AI models, built on its Claude platform, are rivals to industry giants like OpenAI's ChatGPT and Google's Gemini. Fable 5 is a consumer-oriented version of the model capable of deep reasoning and independent complex task performance, while Mythos 5 is designed for businesses and cybersecurity experts, with the ability to identify vulnerabilities in computer code and exploit them. The company had disputed claims that a potential jailbreak should warrant recalling the commercial model deployed to hundreds of millions of people.