Measures of Beijing Municipality on Promoting the Innovative Development of Artificial General Intelligence

• Authority: Chinese provincial and local governments
• Jurisdiction: China
• Collections: Chinese law and policy
• Most recent activity: Proposed
• Most recent activity date: 2023-05-30

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
• Competitive dynamics
• Governance failure
• Lack of capability or robustness

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

• Exposure to toxic content
• Compromise of privacy
• Disinformation, surveillance, and influence at scale
• Cyberattacks, weapon development or use

All other risk subdomains had no coverage in this document.

Risk Coverage Summary

The document has minimal to good coverage of approximately 6-8 subdomains, with primary focus on AI system security (2.2), governance failure prevention (6.5), lack of robustness (7.3), and competitive dynamics (6.4). There is some coverage of malicious actor risks (4.1, 4.2) through security measures, and privacy concerns (2.1). The document emphasizes innovation support, security management, and regulatory frameworks rather than detailed risk mitigation.

Sector coverage

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

• Information
• Finance & Insurance
• Scientific Research & Development Services
• Health Care & Social Assistance
• Public Administration (excl. National Security)

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

• Trade, Transportation & Utilities
• Professional & Technical Services
• Educational Services

All other sectors had no coverage in this document.

Sectors Summary

The document governs AI use across multiple sectors with good coverage of Public Administration (government services, urban governance), Health Care (medical diagnosis and treatment), Finance and Insurance (risk control, advisory services), Scientific Research and Development Services (research collaboration, new materials and drugs), and Information (autonomous driving technology). Additional minimal coverage includes Professional and Technical Services and Educational Services.

Legislative Status

Hard Law

Legislative Status Summary

This is a binding municipal regulation with mandatory language throughout ('shall'), clear enforcement mechanisms including security assessments and algorithm filing procedures, and explicit regulatory oversight by Beijing municipal authorities.

Actors

Proposer

• Entity Types: AI_Governance_Actor
• Entities: Beijing Municipality; Zhongguancun National Independent Innovation Demonstration Zone
• Summary: The document is a municipal regulation issued by Beijing Municipality to implement a previously adopted Implementation Plan for developing Beijing's AI innovation hub.

Target

• Entity Types: AI_Developer, AI_Infrastructure_Provider, AI_Deployer
• Entities: cloud service providers; innovation-driven entities; financial institutions; laboratories
• Summary: The document targets AI developers and deployers in Beijing, including cloud service providers, innovation-driven entities developing AGI technologies, and organizations deploying AI in various sectors like finance, healthcare, and government services.

Enforcer

• Entity Types: AI_Governance_Actor
• Entities: Beijing Municipal Government; relevant regulatory agencies
• Summary: Beijing municipal authorities are responsible for enforcement through security assessments, algorithm filing procedures, and network/data security management oversight, though specific enforcement agencies are not named.

Monitor

• Entity Types: AI_Governance_Actor
• Entities: public service platforms for managing principles and codes of conduct
• Summary: The document establishes monitoring mechanisms including regular service provision, security detection oversight, and public service platforms for managing scientific conduct principles, though specific monitoring bodies are not explicitly named.

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 covers multiple AI lifecycle stages with particular emphasis on Build and Use Model (developing large models, algorithms, evaluation systems), Deploy (applications in various sectors), and Operate and Monitor (security management, ongoing oversight). It also addresses Plan and Design through innovation support and Collect and Process Data through data supply initiatives.

Technical Scope

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

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

Technical Scope Summary

The document explicitly focuses on Artificial General Intelligence (AGI) and large models throughout. It mentions foundational models, multi-modal models, and various AI applications but does not explicitly distinguish between general purpose AI, task-specific AI, or use compute thresholds. There is no explicit mention of generative vs predictive AI, foundation models as a distinct category, or open-weight/open-source models.