Secure agentic AI for your Frontier Transformation

A set of Microsoft Security Blog posts (Feb–Mar 2026) highlights rapidly evolving threats tied to AI and modern application and authentication workflows. Key incidents include malicious browser extensions harvesting LLM/chat histories at scale, Tycoon2FA AiTM phishing-as-a-service reaching hundreds of thousands of organizations, signed malware using a stolen EV certificate to deploy RMM backdoors, OAuth redirection abuse for phishing/malware delivery, and developer-targeting via malicious Next.js repositories that chain RCE to covert C2. Microsoft recommends hardening certificate and RMM/tele‑