Fundamental research in adversarial AI, agent safety, and defense mechanisms
Adversarial Red Teaming