Teardowns, defense, and where AI security is going.
AI is collapsing the cost of producing security work. The scarce thing left is someone who will stake their name on it.
Prompt injection, exfiltration, and jailbreaks are the same bug: the model cannot tell instructions from data.
Guardrails, evals, and threat models for teams actually shipping AI.
New attack cases, and how the frontier models hold up, graded month over month.