Claude Code adds AI-powered security layer

Security has long been seen as one of AI’s biggest pitfalls, and the concerns have only accelerated as agents take on autonomous action. Anthropic is trying to change the narrative.