Quantitative and qualitative metrics to evaluate the effectiveness of AI red teaming activities on artificial intelligence systems and generative models. Includes robustness indicators, adversarial attack success rates, AI attack surface coverage, bias detection metrics, prompt injection success rate, and measurement frameworks for ML/LLM system security.