Anthropic has resumed the limited rollout of its advanced Mythos 5 artificial intelligence model following a decision by the US government to ease restrictions. This move comes after a two-week suspension of access to the model, which was initially imposed due to concerns over national security risks.

The US government has permitted Anthropic to provide access to Mythos 5 to a select group of 'trusted' organizations, signaling a partial reversal of a prior order. This development has been confirmed by a source familiar with the new directive, who noted that more than 100 companies and institutions will now have access to the model.

The decision follows concerns that powerful AI systems could be misused by military intelligence users in countries of concern, prompting the administration to adopt a stringent approach to oversight. This has also affected OpenAI, which has delayed the full public launch of its GPT-5.6 model at the request of US authorities.

The limited access to Mythos 5 is expected to impact the broader AI industry, influencing how companies manage costs, navigate vendor lock-in, and address governance challenges. The move may also affect market dynamics as firms weigh the benefits of restricted access against broader deployment risks.

As the situation continues to develop, the implications of the US government’s decision remain unclear. Anthropic’s ability to provide access to Mythos 5 will likely influence how other AI firms approach regulatory compliance and model deployment in the coming months.