AI Security
What is AI Security?
Measures to protect AI systems from adversarial attacks, data poisoning, model theft, and other threats specific to machine learning systems.
Terms that appear alongside ai security
Each of these is named in at least one of the same controls as ai security. The number is how many controls name both.
- cybersecurity 3 shared controls
- nist 2 shared controls
- owasp 2 shared controls
- data protection 2 shared controls
- governance 2 shared controls
Frameworks that govern ai security
What the standards actually require on ai security
Requirements naming ai security across 4 standards, quoted from the control text.
VEU authorises exports, reexports and transfers of eligible items to end users approved by BIS and listed in Supplement No. 7: General VEU for the PRC and India, and Data Center VEU (universal, for owners headquartered in the AI Authorization countries, or nat...
ear-export-administration-regulations::748.15 · 748.15 Authorization Validated End-User (VEU) →HK PDPO operates under One Country Two Systems separation from mainland China data protection regime (PIPL + DSL + CSL each covered as its own framework).
HK-PDPO-Coord-China-PIPL-APEC-CBPR-2024-2025-Pipeline-AI-Breach · HK PDPO Coordination with China PIPL + APEC CBPR + Global CBPR Forum + 2024 Generative AI Model Framework + Pending 2025 Mandatory Breach + Administrative Fines →Mitigate CBRN information or capabilities risks per NIST AI 600-1 RISK-01 including: training data filtering for chemical + biological + radiological + nuclear hazardous content + model evaluation against WMDP benchmarks + safety filters at inference + uplift...
NISTAI600-6 · CBRN, Cybersecurity, and Information Security Risks (Risks 1, 9) →The 2024 response found that highly capable general-purpose AI challenges a context-based framework and set out a case for targeted binding requirements on the small number of developers of such systems, while relying for now on voluntary measures: the Emergin...
uk-ai-regulation-framework::G1 · Highly capable general-purpose AI: voluntary measures and possible binding requirements →Questions people ask about ai security
What is AI Security?
Why is AI Security important for compliance?
Which compliance frameworks address AI Security?
Where can I learn more about AI Security?
See how AI Security applies across compliance frameworks
Our platform maps 1,016 frameworks with 316K cross-framework control mappings. Explore how this concept is addressed across standards.