CTI-REALM: A new benchmark for end-to-end detection rule generation with AI agents

This collection of Microsoft Security Blog posts announces new AI- and security-focused research, tooling, and investigations. Key items include CTI-REALM, an open-source benchmark for evaluating AI agents that generate end-to-end detection rules from cyber threat intelligence; new guidance and capabilities for securing agentic AI (Zero Trust for AI, observability, and Purview governance enhancements); and practical detection/response guidance for prompt injection and AI observability. Operational threat reporting covers phishing and malware trends—tax-season lures and voice-phishing (Teams v0