NIST adversarial ML guidance
Assessment outcome
total 5/6 (actionability 1 + authority 2 + currency 2)
Eligibility gates
HTTP 200, 14660 words of full text extracted
The source focuses extensively on adversarial machine learning, attack taxonomies, and terminology for AI systems.
It addresses risk management, security, resilience, robustness, and mitigation strategies for AI systems.
Authored and issued by National Institute of Standards and Technology and named researchers.
No indicators or notices on the page state that this document has been superseded, repealed, or withdrawn.
Scored criteria
Actionability
1/2Level 1: The source provides principles and direction regarding mitigations, but no binding steps or specific obligations.
The report also provides corresponding methods for mitigating and managing the consequences of attacks and points out relevant open challenges to take into account in the lifecycle of AI systems.
Authority
2/2Level 2: Official government document published by the National Institute of Standards and Technology.
Approved by the NIST Editorial Review Board on 2024-01-02 How to cite this NIST Technical Series Publication
Currency
2/2computed: newest date 2025-03-24 (via pdf-meta), ~1.4y old; within the 36-month standards_frameworks window for perspective 'Standards body' → 2