Latest posts
$ whoami
Corgi Corp - Adversarial AI Research
AI security research at the behavioral ↔ technical boundary.
$ boot --modules behavior_technical_attack
✓ research archive initialized
$ ls ./focus
⠀⠀⠀⠀. ⠀⣰⣧⣼⣧⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣿⣿⣭⣭⣤⣄⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣼⣿⣿⣿⣿⣿⣷⣤⣤⡄
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣿⣿⣿⣿⣿⣿⣿⣿⡿⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣀⣼⣿⣮⣍⣉⣉⣀⣀⠀⠀⠀
⠀⠀⣠⣶⣶⣶⣶⣶⣶⣶⣶⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣧⠀⠀
⣴⣿⣿⣿⣿⣿⣯⡛⢿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⠀⠀
⠉⣿⣿⣿⣿⣿⣿⣷⠘⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⠏⠀⠀
⠀⣿⣿⣿⣿⣿⣿⡟⠸⠿⠿⠿⣿⣿⣿⣿⣿⣿⣿⠿⠋⠀⠀⠀
⠀⠘⢿⣿⣿⠿⠋⠀⠀⠀⠀⠀⠀⠉⠉⣿⣿⡏⠁⠀⠀⠀⠀⠀
⠀⠀⢸⣿⣿⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣿⣿⡇⠀
_
Featured research
Context Is Not Identity: Why AI Security is an Authorization Problem
AI security isn’t a prompt problem; it’s an authorization problem. Learn how identity, capability, and enforcement secure agentic AI systems.
Read analysis →