NIST AI Risk Management Framework Quiz

Answers and explanations appear as you take the quiz.

1. What are the four core functions of the NIST AI RMF?
   • Govern, Map, Measure, Manage
   • Plan, Build, Test, Deploy
   • Detect, Protect, Respond, Recover
   • Collect, Train, Validate, Ship
2. Is compliance with the NIST AI RMF legally required?
   • Yes, it is binding US federal law
   • No, it is a voluntary framework
   • Only for companies based in Europe
   • Only for AI used by the military
3. Which function is described as cross-cutting and feeding into the other three?
   • Map
   • Measure
   • Govern
   • Manage
4. Why does the RMF treat trustworthiness as a set of trade-offs?
   • Because all AI properties always improve together
   • Because only accuracy matters in practice
   • Because trade-offs are illegal to consider
   • Because improving one property can degrade another
5. What did NIST add to the framework in 2024 to address large language models?
   • A Generative AI Profile
   • A total ban on chatbots
   • A licensing fee schedule
   • A replacement framework called AI RMF 2.0
6. Which of these is one of NIST's named characteristics of trustworthy AI?
   • Maximally profitable
   • Explainable and interpretable
   • Fastest possible inference
   • Built only with open-source code
7. Which risk does NIST’s Generative AI Profile identify as relevant to generative-AI systems?
   • A requirement to use open-source weights
   • A guarantee that the model will memorize no data
   • Confabulation, in which generated content may be presented as factual
   • A rule that all generated images must be watermarked
8. What does the NIST AI RMF Playbook provide?
   • Legally binding rules for all organizations
   • A guarantee that a model is safe
   • A replacement for risk measurement
   • Suggested actions and references to help achieve framework outcomes