NIST standardizes the attack language for adversarial machine learning
NIST published AI 100-2e2025, a voluntary taxonomy covering evasion, poisoning, privacy and misuse attacks across predictive and generative AI. The report maps attacker goals, capabilities and mitigations while explicitly noting where defenses remain incomplete.
Why it matters: A common vocabulary lets security, model and governance teams threat-model the same system without collapsing very different failure modes into the word “jailbreak.”