AI Dynamics

Global AI News Aggregator

About

Defenses tested and how they died: prompting, training, filtering

Here's every defense tested and how they died: • Prompting: Spotlighting, Prompt Sandwich, RPO → Killed by Search & RL
• Training: Circuit Breaker, StruQ, MetaSecAlign → Killed by RL
• Filtering: ProtectAI, PromptGuard, PIGuard, Model Armor → Killed by Search & Humans

→ View original post on X — @godofprompt