US allows Anthropic to release Mythos AI to 'trusted' organisations

The U.S. government has granted Anthropic permission to release its powerful Claude Mythos 5 artificial intelligence model to a group of "trusted" U.S. organizations. This move partially rescinds an earlier directive that had suspended access to the AI over national security risks.

The decision means that more than 100 companies and institutions, many of them Fortune 500 corporations, will now gain access to Mythos 5. This information comes from a source familiar with the new directive who spoke on condition of anonymity due to the sensitive nature of the matter.

The U.S. administration has been taking a proactive stance on overseeing the release of advanced AI models from companies like Anthropic and its rival OpenAI. This approach stems from concerns that sophisticated AI systems could be exploited by military intelligence entities in countries identified as potential adversaries, such as China and Russia.

This development follows a similar announcement from OpenAI, which stated it was delaying the full public launch of its GPT-5.6 model at the U.S. government's request, restricting its availability to a limited group.