When the U.S. government forced Anthropic to pull its most powerful AI models offline on June 13, 2026, it transformed abstract policy concerns into a geopolitical reality. By mandating an immediate suspension of access for all foreign nationals, Washington inadvertently signaled to global allies that the world’s critical AI infrastructure remains tethered to American political whims.
The directive, issued by Commerce Secretary Howard Lutnick to CEO Dario Amodei, targeted the Claude Fable 5 and Mythos 5 models. Anthropic, unable to instantly restrict access by nationality, shuttered the services entirely. The administration cited national security risks, specifically a method for bypassing safety guardrails, which David Sacks of the President’s Council of Advisers on Science and Technology claimed Anthropic refused to patch. Amazon CEO Andy Jassy reportedly fueled the decision, warning officials that the models could be leveraged to facilitate cyberattacks.This move marks the escalation of a long-standing confrontation between the Trump administration and Anthropic. Earlier this year, the Pentagon designated the company a "Supply-Chain Risk to National Security" after it resisted military applications for its technology, leading to a federal ban on its use. Critics view the latest export controls as inconsistent, particularly as the U.S. simultaneously permits the export of advanced AI hardware to China.
The fallout has rippled across Europe and Canada, where officials are questioning the risks of relying on American-controlled AI. European Commission spokespeople and national leaders, including Finnish MEP Aura Salla, have criticized the move as discriminatory. Canadian Prime Minister Mark Carney described the incident as a stark lesson in the dangers of over-reliance on a handful of U.S. providers. As governments scramble to assert technological autonomy, the episode has solidified a new consensus: access to frontier AI is no longer merely a commercial agreement, but a volatile function of U.S. jurisdiction.

Comments (0)
No comments yet. Be the first!