Anthropic launches Claude Security to counter rapid AI-Powered exploits

Anthropic has launched Claude Security (public beta for enterprise) — an AI-driven code-scanning and remediation tool built on Claude Opus 4.7 — positioned to help defenders keep pace with rapidly improving AI-powered offensive tools (e.g., Mythos). The announcement highlights features such as targeted repository scans, confidence scoring, integration with vendors and workflow tools, scheduled scans, and export/webhook capabilities, and frames the product as a defensive response to a landscape where AI can drastically shorten time-to-exploit.