NIST-AI-RMF

NIST AI RMF AI safety + governance framework.

Définition

NIST AI RMF (Artificial Intelligence Risk Management Framework) est le voluntary framework publie janvier 2023 par US NIST (National Institute of Standards and Technology) pour aider organisations gerer AI risks throughout AI lifecycle, 4 core functions (GOVERN + MAP + MEASURE + MANAGE) + AI Risk Management Framework Playbook accompagnante + AI RMF 1.0 Generative AI Profile published juillet 2024 specific GenAI guidance, foundational US AI governance approach voluntary alignement private sector + government adoption. Framework + standard + guidance objectifs AI safety + governance + risk management + transparency + fairness + accountability + explainability + multiple AI ethics principles adoption multi-stakeholder process + voluntary adoption industry + emerging regulation alignment EU AI Act + US Executive Orders + multiple national jurisdictions + private sector adoption Microsoft + Google + Meta + OpenAI + Anthropic + autres major AI labs + corporate compliance teams 2020s+.

Origine

NIST AI RMF 1.0 published janvier 2023 ; AI RMF 1.0 Generative AI Profile published juillet 2024 ; widely adopted US private sector + government voluntary basis.

Exemple en contexte

US Fortune 500 corporation deploys AI customer service chatbot : implements NIST AI RMF GOVERN function (AI governance policies + roles + responsibilities) + MAP function (identify AI system context + impacts + risks bias + privacy + security) + MEASURE function (assess + monitor AI performance + risks) + MANAGE function (mitigate identified risks + incident response procedures) ensuring responsible AI deployment.

Termes liés

• ISO/IEC 23894 — international AI risk standard.