Artificial Intelligence Security Standardization White Paper, Section 5 ("Recommendations for AI security standardization work")

• Authority: Other authorities
• Collections: Chinese law and policy
• Most recent activity: Proposed
• Most recent activity date: 2024-11-01

Risk subdomain coverage

Good coverage — Comprehensive governance measures or mitigations with clear procedures relating to the risk subdomain are explicitly described.

• AI system security vulnerabilities
• Governance failure
• AI possessing dangerous capabilities
• Lack of capability or robustness

Minimal coverage — Brief mention but no specific mitigations, recommendations or governance measures relating to the risk subdomain are described.

• Unfair discrimination and misrepresentation
• Unequal performance across groups
• Compromise of privacy
• Disinformation, surveillance, and influence at scale
• Cyberattacks, weapon development or use
• Fraud, scams, and targeted manipulation
• Competitive dynamics
• AI pursuing its own goals
• Lack of transparency or interpretability

All other risk subdomains had no coverage in this document.

Risk Coverage Summary

The document has good coverage of approximately 10-12 subdomains, with strong focus on AI system security (2.2), malicious actors and cyberattacks (4.1, 4.2), governance failure (6.5), competitive dynamics (6.4), and AI safety failures (7.1, 7.2, 7.3, 7.4). Coverage is concentrated in security, governance, and AI system robustness domains.

Sector coverage

Good coverage — Comprehensive governance measures or mitigations with clear procedures relating to the sector are explicitly described.

• Information
• Public Administration (excl. National Security)

Minimal coverage — Brief mention but no specific mitigations, recommendations or governance measures relating to the sector are described.

• Agriculture, Mining, Construction & Manufacturing
• Trade, Transportation & Utilities
• Finance & Insurance
• Professional & Technical Services
• Scientific Research & Development Services
• Educational Services
• Accommodation, Food & Other Services
• National Security

All other sectors had no coverage in this document.

Sectors Summary

The document governs AI security across multiple critical infrastructure sectors including telecommunications, energy, transportation, power, and finance. It also addresses manufacturing (smart manufacturing, intelligent networked vehicles) and consumer products (smart homes, smart door locks). The governance is cross-sectoral with emphasis on supply chain security management.

Legislative Status

Soft Law

Legislative Status Summary

This is a white paper providing recommendations for AI security standardization work, using voluntary language throughout ('we recommend', 'should be') without binding legal obligations or enforcement mechanisms.

Actors

Proposer

• Entity Types: AI_Governance_Actor
• Summary: The document is a white paper providing recommendations for AI security standardization work in China, with references to government planning and national standardization efforts. The National Information Security Standardization Technical Committee is explicitly mentioned as having carried out standards development work.

Target

• Entity Types: AI_Developer, AI_Deployer, AI_Governance_Actor
• Summary: The document targets multiple stakeholder groups involved in AI development, deployment, and governance. It explicitly addresses enterprises developing AI products, government agencies implementing supervision, and organizations across critical infrastructure sectors.

Enforcer

• Entity Types: AI_Governance_Actor
• Summary: The document recommends that government establish monitoring and supervision systems for AI security, though specific enforcement agencies are not named. References to government supervision and regulatory frameworks indicate governmental enforcement role.

Monitor

• Entity Types: AI_Governance_Actor
• Summary: The document establishes monitoring mechanisms through pilot programs and evaluation systems, with multiple stakeholders involved in tracking standards implementation and effectiveness.

AI Lifecycle Stages

Lifecycle stages covered by the documents, as defined in the OECD Framework for the Classification of AI Systems

• Plan and Design
• Collect and Process Data
• Build and Use Model
• Verify and Validate
• Deploy
• Operate and Monitor

AI Lifecycle Summary

The document comprehensively covers all AI lifecycle stages, with particular emphasis on the Plan and Design stage (ethical principles, security architecture), Build and Use Model stage (algorithm security, model trustworthiness), and Operate and Monitor stage (ongoing supervision, supply chain management). It addresses the entire lifecycle from initial planning through deployment and operational monitoring.

Technical Scope

Terms relating to technical scope that are mentioned or implicitly referenced in the document

• AI Models
• AI Systems
• Open Weight / Open Source

Technical Scope Summary

The document broadly addresses AI systems, models, algorithms, and products without defining specific technical categories. It mentions AI technologies, products, and applications generally, with references to algorithm models, smart products, and open source frameworks. No specific compute thresholds, frontier AI, or GPAI definitions are provided.