Bottom line: Anthropic is opening its cyber-capable Mythos model to approximately 150 new organizations across more than 15 countries, including EU member states for the first time, with strict security requirements.
Anthropic is granting approximately 150 new organizations from over 15 countries access to its Mythos AI model, which can exploit software vulnerabilities. The expansion now includes EU countries such as France, Germany, and Italy.
The California-based AI company Anthropic is expanding its Glasswing Project, a coalition of trusted technology and cybersecurity firms with access to the Mythos model. This coalition was originally founded with organizations primarily from the US to develop defensive mechanisms against cyberattacks. With the new expansion, approximately 150 additional organizations from more than 15 countries now gain access.
The affected EU countries are France, Germany, Italy, Switzerland, the Netherlands, Spain, Belgium, and Sweden. Anthropic announced it will expand Glasswing “even further,” stating that future expansions will encompass organizations in the US and abroad equally. However, each new organization must meet Anthropic’s security requirements before gaining access.
The announcement follows Anthropic’s invitation to the European cybersecurity agency ENISA to use the model following a meeting in San Francisco. The EU Commission must now establish a mechanism to provide secure access. European regulators were initially not informed about the model’s capabilities. The expansion demonstrates how Anthropic, despite initial lack of transparency toward EU authorities, is attempting to create structured access to its cyber capabilities — while taking into account the regulatory requirements of the EU AI Act.
Source: www.politico.eu · Published June 2, 2026
Lumi AI News — AI-assisted curation pursuant to Article 50 EU AI Act. Paraphrase and classification by Lumi News Pipeline v1.2.9.