Anthropic confirms it is collaborating with other companies on a project called Glasswing, in an effort to safeguard critical software from AI-driven
exploits, Bleeping Computer writes. This occurs as Anthropic prepares for the public rollout of Mythos, a model that it says offers improvements in cold reasoning and autonomy far
above its current flagship model, Opus 4.7. The model may be coming to Claude Code. However, Mythos can automatically develop functional cyberattacks at a highly professional
level. “The advantage will belong to the side that can get the most out of these tools,” Anthropic warns.