Anthropic launched “Project Glasswing,” giving major technology and cybersecurity firms access to a preview version of its unreleased Claude Mythos model to strengthen defenses against AI-enabled cyberattacks. The company says it is concerned that its most capable model could increase the likelihood of large-scale, agentic cyber incidents. Under the initiative, organizations including AWS, Apple, Cisco, CrowdStrike, Google, JPMorgan Chase, Microsoft, and NVIDIA can use the Mythos Preview for defensive security work, including scanning systems and open-source code for vulnerabilities. Anthropic says the model will not be generally available, but it aims for safe deployment at scale once additional safeguards are in place. The initiative also follows earlier market concerns tied to reports about Anthropic developing Mythos-class cyber capabilities, and it signals how campus and research communities will increasingly face AI toolchains where cybersecurity and model governance are bundled into deployment decisions.