Anthropic disabled its most advanced AI models worldwide after a US export control directive, triggering emergency talks between European and American leaders over a 'trusted partners' framework.
Anthropic pulled the plug on global access to its two most powerful AI models after the Trump administration issued an export control directive citing national security risks.
On June 12, Commerce Secretary Howard Lutnick delivered a letter mandating that Anthropic block all foreign nationals from accessing its Fable 5 and Mythos 5 models. Rather than try to build a system that selectively filters users by nationality, Anthropic chose the nuclear option: it disabled the models entirely.
The fallout and the scramble #
The decision didn’t just affect adversarial nations. It cut off allied governments, European businesses, and research institutions that had integrated Anthropic’s technology into their workflows.
The EU wasted no time raising the issue directly with the Trump administration. French President Emmanuel Macron emerged as the most vocal European leader pushing for a resolution, using the G7 summit in Evian as a platform to advocate for what he called a “trusted partners” framework. The idea is straightforward: create a vetted list of allied nations whose citizens and institutions can access advanced US AI systems without triggering national security alarms.
Anthropic’s own leadership got involved in the diplomatic effort. Co-founder Tom Brown engaged in negotiations with US officials, including National Cyber Director Sean Cairncross, to find a path back to restored access.
The directive reportedly stemmed from concerns about potential jailbreak vulnerabilities in Fable 5 and Mythos 5, risks that could allow bad actors to circumvent the models’ safety guardrails.
Europe’s AI sovereignty push gains momentum #
The Anthropic shutdown has turbocharged existing calls from European leaders for greater AI sovereignty. The argument is that depending on US companies for foundational AI capabilities creates an unacceptable strategic vulnerability.
Macron’s push for a “trusted partners” framework represents a pragmatic middle ground. Full AI independence is a multi-year, multi-billion-dollar project. Getting back on the approved list for US models is something that could theoretically happen in weeks or months.
What this means for investors #
The immediate market implications center on uncertainty. Companies that built products or services on top of Anthropic’s Fable 5 and Mythos 5 models just had the rug pulled out from under them, with no clear timeline for restoration. Any business with meaningful exposure to Anthropic’s API outside the US is now dealing with a supply chain disruption for intelligence itself.
Investors should also pay close attention to how Anthropic handles this internally. The company chose full compliance over selective restriction, which signals either strong alignment with US government priorities or a lack of technical capability to implement granular access controls.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our