A two-stage exploit chain in vLLM's multimodal video processing bypasses ASLR through PIL exception leakage, then achieves heap overflow via a malicious JPEG2000 file, giving unauthenticated attackers code execution on inference servers.
AI agents generate a new class of security event that existing SIEM infrastructure was not designed to process. DeepMind's June 2026 control roadmap, OWASP's Agentic Top 10, and the EU AI Act's August logging deadline all converge on the same problem: security teams cannot monitor what they are not capturing.
CVE-2026-12957 lets a malicious repository silently execute arbitrary commands the moment a developer opens it in Amazon Q Developer, exfiltrating AWS credentials with no user interaction required.
Three papers published in 2026 confirm what practitioners suspected: LLM safety alignment is structurally shallow, and fine-tuning APIs are the widest open bypass.
Zenity Labs has documented attackers exploiting unpatched LiteLLM vulnerabilities and unauthenticated Ollama endpoints to run autonomous pentesting campaigns, steal compute, and exfiltrate data from victim organisations.