The United States government has partially reversed its earlier restrictions on Anthropic’s advanced AI system, Mythos 5, allowing a carefully selected group of trusted American cybersecurity organizations and critical infrastructure operators to regain access. The decision follows a June 12 order by the Trump administration that halted public deployment of both Mythos 5 and Fable 5 over national security concerns.
Officials had argued that highly capable frontier AI models could potentially be exploited by military or intelligence actors from rival nations, including China and Russia, prompting tighter federal oversight of powerful artificial intelligence technologies. The restrictions sparked criticism from technology leaders and policy observers who viewed the measures as excessive government intervention in AI innovation.
Anthropic confirmed that more than 100 companies and institutions, including major Fortune 500 organizations, are expected to benefit from the new directive. However, authorities have not disclosed the criteria for selecting eligible recipients or the process through which access approvals will be granted.
The move has reignited debate within the tech industry about the balance between national security and open technological development. OpenAI CEO Sam Altman publicly supported robust safety evaluations but questioned the principle of government agencies determining which organizations may use cutting-edge AI systems.
