Comprehensive AI / LLM Security Guide
Comprehensive AI / LLM Security Guide 🆕 Updated September 12, 2026 — added the LiteLLM gateway chain (CVE-2026-59821/59822): default keys to RCE to IAM theft, and why the SSRF in it will never get a CVE. A practitioner’s reference for securing Large Language Model and agentic AI systems — attack surface, exploitation techniques, real-world CVE chains, payloads, and layered detection/prevention. Draws on published research from OWASP, NVIDIA AI Red Team, Unit 42, Lakera/Check Point, NCSC, CrowdStrike/Pangea, Equixly, Anthropic, OpenAI, Microsoft MSRC, Google, AWS, MITRE ATLAS, Red Hat, Pillar Security, JFrog, AuthZed, and Trend Micro. ...