EU cybersecurity agency granted access to Anthropic's Mythos 5 model for testing

The European Commission has confirmed that the European Union Agency for Cybersecurity (ENISA) has been granted official access to Anthropic's restricted-access Claude Mythos 5 model.

The high-tier AI model, recognized for its advanced capabilities in tracking and discovering deep software vulnerabilities, is being made available to evaluate emerging digital defense risks.

Because the Mythos class operates with reduced cyber safeguards compared to standard public models, its deployment is strictly controlled under vetted research and defensive security initiatives.

European regulators aim to leverage the model’s deep threat-mapping architecture to stress-test regional critical infrastructure and understand complex autonomous exploit mechanics.

This development follows intense global scrutiny over the export and deployment of frontier-class artificial intelligence systems.

By integrating the advanced security model into ENISA's testing pipelines, the European Union seeks to close the capability gap between fast-evolving generative tech and institutional cybersecurity preparedness.