A Chinese threat group has pushed the boundaries of cybercrime by turning an AI model into a fully autonomous hacking engine.
Anthropic said this is the first documented case of a large-scale cyberattack executed with minimal human intervention.