NIST’s ARIA program develops sociotechnical evaluation methods that test AI through model testing, red-teaming, and field testing, scoring systems on a Contextual Robustness Index to show how they behave under varied real-world conditions.
Tags
assessmentevaluationNISTARIAsociotechnical
At a glance
Published
2024
Jurisdiction
United States
Category
Assessment and evaluation
Access
Public access
Build your AI governance program
VerifyWise helps you implement AI governance frameworks, track compliance, and manage risk across your AI systems.