Leaders from the Group of Seven (G7) nations convened in Evian-les-Bains, France, to discuss a proposal to grant selected "trusted partners" access to advanced artificial intelligence models developed by major US technology companies, including Anthropic. This discussion emerged after the US government, under President Donald Trump, ordered Anthropic to restrict foreign national access to its most advanced AI models, such as Fable 5 and Mythos 5, citing national security concerns. As a result, Anthropic disabled access to these models for all users, including those in countries where it had previously provided access to organizations in various sectors.
This incident has intensified concerns among G7 allies, particularly in Europe, about the potential for a US "kill switch" that could unilaterally cut off access to critical American technology. The restrictions highlighted a growing push for "digital sovereignty" among European countries, with the European Union expressing interest in access to models like Mythos to evaluate their risks. Cybersecurity experts have also voiced worries about the capabilities of Anthropic's Mythos model, which is designed to identify weaknesses in computer code, and its potential misuse for cyberattacks.
The G7 discussions focused on developing a framework to expand AI access while addressing US security concerns. Representatives from G7 nations raised the idea of broader access during meetings with US officials. The proposed "trusted partners" could be governments or companies. AI company CEOs, including Dario Amodei from Anthropic, Sam Altman from OpenAI, and Demis Hassabis from Google DeepMind, were scheduled to join G7 discussions later in the week to address issues such as AI regulation and infrastructure. Canadian Prime Minister Mark Carney specifically noted the Anthropic incident as highlighting the need to "build out and diversify" AI access.
The United Kingdom, through a No. 10 spokesperson, denied media reports that it had sought a carve-out from the US export controls on Anthropic, stating such claims were "categorically untrue." The spokesperson affirmed that the UK is in regular contact with both the US government and Anthropic to understand the situation. The episode has underscored a broader sentiment among Western allies that relying solely on US technology carries risks, pushing countries to seek greater independence in their AI ecosystems.
