US Government to Vet Access to Latest American AI Technology from OpenAI and Anthropic

The Trump administration has introduced a significant shift in its approach to artificial intelligence, now requiring government approval for customers seeking access to the most advanced AI technologies from leading American developers like OpenAI and Anthropic. This new policy aims to ensure national security amidst growing concerns over the powerful capabilities of frontier AI models. Specifically, OpenAI announced that the U.S. government would initially approve who gains access to its latest releases, including the new GPT-5.6 models (Sol, Terra, and Luna). This measure is described as a short-term step while the administration and AI companies work towards a more comprehensive regulatory framework. Similarly, the Commerce Department informed Anthropic that its advanced AI model, Mythos 5, would only be accessible to a restricted list of U.S.-based companies. The administration's stance has evolved rapidly, moving from a previously hands-off approach to a more interventionist one, particularly after the emergence of AI systems capable of identifying software vulnerabilities. This has sparked debate within the AI community, with some, like OpenAI CEO Sam Altman, expressing reservations about increased federal oversight, while others acknowledge the need for caution. Critics argue that such stringent controls could hinder innovation and prevent trusted partners, including cybersecurity defenders, from accessing crucial tools. However, the government emphasizes the need to prevent critical technologies from falling into the wrong hands or being exploited by adversarial nations. The policy reflects a delicate balancing act between fostering technological advancement and mitigating potential risks associated with increasingly powerful AI.