ENISA, the European Union's cybersecurity agency, has gained access to Anthropic's Mythos model, a large-scale artificial intelligence system designed to assist with complex tasks. In May, Anthropic, a U.S.-based artificial intelligence company, granted access to a model known as GPT-5.5-Cyber but withheld access to its most advanced cybersecurity-focused model. A month later, Anthropic announced that it would expand access to the Mythos model to government organizations outside the United States, including the European Union, though this was not officially confirmed at the time. According to a representative speaking to Bloomberg, the European Commission has confirmed that ENISA has access to Mythos 5 and is currently testing the system. However, Anthropic has not officially commented on the matter. It remains unclear whether ENISA has access to the most recent version of the Mythos model. Last week, Anthropic announced the release of Mythos version 5.1, an updated iteration of the model. Dario Amodei, co-founder of Anthropic, stated that this version is accessible only to a select number of American organizations, suggesting that access to the latest version may still be limited to specific entities. The expansion of access to the Mythos model highlights the growing interest in leveraging advanced AI systems for cybersecurity purposes. As governments and organizations explore the potential of such models, questions remain about the extent of access and the implications for international collaboration in the field of digital security.