On April 7, 2026, Anthropic announced Claude Mythos Preview, a frontier model designed for advanced cybersecurity tasks. Through Project Glasswing, the model was made available to select launch partners and organizations supporting critical software infrastructure. This week, a highly capable model from Z.ai called Ox Alpha was released with open weights -- representing the democratization of frontier cyber-offensive capabilities.  As the Five Eyes stated in June -- AI models like Ox Alpha will accelerate "the speed, scale and sophistication of cyber threats.... shrinking the window between vulnerability discovery and exploitation ever more quickly."

OpenAI, Anthropic and other major companies have called on leaders across industry and government in an open letter to bring the full weight of their technology, resources, and expertise to put cyber-capable AI in the hands of defenders.

The Media & Entertainment industry must transition from isolated defense to a coordinated, real-time threat-sharing framework with verified identity for secure collaboration and best practice and skills development. The GC-SC has launched Collective Defense for Media & Entertainment -- a threat- and intel-sharing program built in partnership with the GRF (Global Resilience Foundation). By partnering with GRF, GC-SC gains the ability to rapidly build and scale an M&E-specific platform, while maintaining exposure to broader cross-industry initiatives through the Business Resilience Council.