Articles
Cisco AI Defense Policy Studio: Turning Unwritten Policy into Adaptive AI Guardrails
5 min read
Cisco's Integrated AI Security and Safety Framework and our recent work on defining taxonomy constitutions focused on defining and detecting common risks shared among enterprises when deploying AI. However, while most enterprises share a lot of.....
Improving Labeling Consistency with Detailed Constitutional Definitions and AI-Driven Evaluation
5 min read
Enterprises need to know exactly what their systems detect, and that definition must stay consistent over time. Writing a definition precise enough to settle every hard case has long been impractical because human annotators cannot hold a document..