Anthropic’s advanced AI models, Mythos 5 and Fable 5, are now available to users outside the United States after the US government lifted restrictions imposed in June. The Trump administration initially blocked access to these “super-AI” models for foreign entities due to concerns they could be exploited for national security risks, potentially enabling mass surveillance or cyberattacks. These concerns stemmed from a tip suggesting the AI’s safeguards could be bypassed. Anthropic protested the initial ban, arguing the response to a limited risk didn't warrant recalling a widely-used commercial model, hence impacting all users globally. Following discussions with the US government, and reinforcement of security measures, the restriction has been partially lifted. Mythos is now accessible to specific US government agencies, and wider international access has been restored, signifying a compromise between commercial interests and national security concerns.