Anthropic shuts the EU out of its most advanced cyber AI model (www.theparliamentmagazine.eu)

Anthropic, a California-based AI company, has restricted access to its advanced cybersecurity model, Claude Mythos, for the European Union, granting use only to selected U.S. companies and government agencies. This decision raises significant concerns for Europe, as it threatens to heighten cybersecurity risks and deepen reliance on American technology providers. The EU has sought access to Mythos to evaluate and bolster its defenses against cyber threats, especially for critical sectors like banking, but to no avail, while U.S. competitors receive early access. This move highlights a widening gap in global AI access, where key technologies are available in the U.S. before Europe due to strict regulatory frameworks. Lawmakers in the EU have voiced frustration, emphasizing the need for Europe to develop its own cutting-edge AI capabilities. The situation reflects broader geopolitical tensions over AI sovereignty and security, especially as Anthropic faces scrutiny from the U.S. government regarding military applications of its models. The EU's ability to respond to potential cybersecurity threats produced by such powerful tools is consequently hindered, raising fundamental questions about its digital sovereignty and dependency on external technology providers.