Secure By Dezign
Pax
Secure By Dezign is the AI Security Training Ground — a daily technical podcast for CISOs, security architects, and AI pentesters who refuse to be caught flat-footed by adversarial machine learning.Every episode dissects a real AI attack technique from first principles: the root vulnerability, step-by-step exploitation with working code, historical breaches, and actionable defenses you can deploy today. Topics include prompt injection, RAG poisoning, model inversion, adversarial ML, LLM jailbreaking, indirect prompt injection, agentic AI security, AI supply chain attacks, deepfake-powered BEC, zero-trust for ML infrastructure, and much more.Whether you're briefing a board, hardening a production LLM deployment, or building your own attack lab — Secure By Dezign gives you the technical depth to do it right.New episodes every weekday. No fluff. Just signal.