Ethical Norms for New Generation Artificial Intelligence
China · Guideline · National Committee for New Generation AI Governance
Our summary
A non-binding ethical guidance document issued by China's AI governance committee that integrates ethical principles throughout the entire lifecycle of artificial intelligence. It applies to individuals, organizations and institutions engaged in AI management, research, supply and deployment activities, and sets out six basic ethical requirements plus eighteen specific requirements across these four activity areas.
Written by the Index in our own words. It is not legal advice; the official text is the authority.
Inside this policy
Every section of this policy in order. Our notes summarise each one in plain language and do not quote the official text, which is the authority.
Drafted by the Index An automated summary of the official text, not yet reviewed by a person.Reviewed Checked by a person against the official text.
31 sections
Jump to a chapter
第一章 General provisions
4 sections
This chapter contains foundational principles for AI ethics, including the scope of the guideline, its core objectives, and basic ethical standards that apply across all AI activities.
第一条 Objective of guideline
The guideline aims to embed ethical values throughout AI development and operation, to advance fairness, justice, harmony and safety, and to prevent bias, discrimination, privacy violations and information breaches.
第二条 Scope of applicability
This guideline applies to all natural persons, legal entities and relevant institutions engaged in AI management activities (strategy, policy, standards, resource allocation, oversight), research activities (science, technology development, product development), supply activities (production, operation, sales), and use activities (procurement, consumption, operation).
第三条 Basic ethical standards
All AI activities must follow six basic ethical standards: advancing human welfare by respecting rights and promoting sustainable development; promoting fairness by ensuring inclusive benefits and supporting disadvantaged groups; protecting privacy by processing personal information lawfully and preventing unlawful collection; ensuring controllability by maintaining human autonomy and the right to refuse or stop AI services; strengthening accountability as humans remain the final responsible party; and enhancing ethical competence through learning and dialogue.
第四条 Activity-specific standards
Specific AI activities must comply with separate ethical standards covering management, research, supply and use activities.
第二章 Management standards
5 sections
This chapter establishes ethical standards for entities engaged in AI management activities, including strategic planning, policy-making and oversight.
第五条 Promote agile governance
AI managers should respect development patterns and recognise both potential and limits of AI, continuously refining governance mechanisms, and advancing AI in an orderly way without haste or unrealistic expectations.
第六条 Actively practice demonstration
Managers must comply with AI laws and standards, integrate ethics throughout management, lead by example in ethical governance, and proactively address public ethical concerns about AI.
第七条 Exercise power correctly
Managers must define the scope and limits of their AI-related authority, regulate how power is used, fully protect the rights and lawful interests of individuals and organisations, and prohibit abuse of power that harms others.
第八条 Strengthen risk prevention
Managers should maintain risk awareness, assess potential AI risks, conduct ongoing risk monitoring and evaluation, establish early warning mechanisms, and strengthen their capacity to manage and respond to AI ethical risks.
第九条 Promote inclusive openness
Stakeholders should recognize the rights and interests of all parties involved in artificial intelligence, support diverse applications to address economic and social challenges, encourage collaboration across disciplines and borders, and work toward widely agreed governance frameworks and standards.
第三章 Development standards
4 sections
This chapter covers standards for artificial intelligence research and development, including requirements for self-discipline, data quality, system safety and transparency, and avoidance of bias and discrimination.
第十条 Strengthen self-discipline
Organizations conducting artificial intelligence research must integrate ethical principles throughout development stages, conduct internal reviews, strengthen self-management, and refrain from developing systems that violate ethical norms.
第十一条 Improve data quality
At all stages of data collection, storage, use, processing, transmission, provision and disclosure, organizations must comply with applicable data laws, standards and rules while enhancing completeness, timeliness, consistency, standardization and accuracy.
第十二条 Enhance safety and transparency
During algorithm design, implementation and application, developers must increase transparency and explainability, improve system reliability, controllability, resilience and interference resistance, and work toward systems that are verifiable, auditable, supervizable, traceable, predictable and trustworthy.
第十三条 Avoid bias and discrimination
During data collection and algorithm development, organizations must conduct ethics reviews, consider diverse needs and perspectives, prevent data and algorithmic bias, and strive to achieve fairness and non-discrimination in artificial intelligence systems.
第四章 Supply standards
4 sections
This chapter covers standards for supplying and deploying artificial intelligence products and services, including requirements for respecting market rules, managing quality, and protecting user rights.
第十四条 Respect market rules
Organizations must comply with market entry, competition and transaction regulations, maintain market order, and are prohibited from using data monopolies or platform monopolies to disrupt competition or infringing others' intellectual property through any means.
第十五条 Strengthen quality control
Organizations must monitor product and service quality and evaluate usage, prevent harm to safety, property and privacy from design or product defects, and must not distribute, sell or supply products and services that fail to meet quality standards.
第十六条 Protect user rights
When using artificial intelligence in products and services, organizations must disclose this to users, indicate functions and limitations, guarantee users' right to be informed and consent, provide straightforward options to choose or exit artificial intelligence modes, and must not create barriers to equal access.
第十七条 Emergency preparedness measures
Developers and deployers must establish emergency mechanisms and loss compensation schemes, continuously monitor AI systems, respond to user feedback promptly, prevent systematic failures, and be ready to assist relevant parties with lawful intervention in AI systems to minimise harm and mitigate risks.
第五章 Usage standards
5 sections
This chapter sets out standards for the use of AI products and services, covering beneficial use, avoiding misuse and abuse, prohibiting unlawful use, providing feedback on problems, and building competence to use AI systems safely and effectively.
第十八条 Promoting beneficial use
Users are encouraged to conduct thorough evaluation and assessment before deploying AI products and services, understand their benefits, and consider the legitimate interests of all stakeholders in order to support economic growth, social advancement and sustainable development.
第十九条 Avoiding misuse and abuse
Users must understand the applicable scope and potential negative effects of AI products and services, respect the right of relevant parties not to use them, prevent inappropriate or abusive use, and avoid unintentional harm to the legitimate rights of others.
第二十条 Prohibiting unlawful use
Users are prohibited from deploying AI products and services that do not comply with laws, regulations, ethical principles or technical standards, or using them for illegal purposes, endangering national security, public safety or production safety, or harming public interests.
第二十一条 Timely issue reporting
Users must actively participate in AI ethics governance, promptly report to relevant parties any technical security vulnerabilities, regulatory gaps, or oversight delays discovered during use of AI products and services, and assist in resolving such issues.
第二十二条 Enhancing user competence
Users should proactively learn about AI, acquire skills needed for operating, maintaining and managing emergencies with AI products and services across all usage stages, and ensure their safe and efficient use.
第六章 Implementation and organisation
3 sections
This chapter addresses how the guideline is administered and implemented, including designation of responsibility for release and interpretation, and provision for other organisations to develop supplementary ethical standards tailored to their specific needs.
第二十三条 Release and interpretation
The National AI Governance Expert Committee is responsible for releasing this guideline and providing interpretation and implementation guidance.
第二十四条 Supplementary standards
Government agencies, enterprises, universities, research institutions, associations and other bodies may develop more specific ethical standards and measures based on this guideline and their particular requirements.
第二十五条 Entry into force and revision
The guidelines take effect from the date of publication and will be revised as needed in response to economic and social development requirements and developments in artificial intelligence.
Controls
What this policy requires, forbids, grants or penalises, one control at a time and in our own words. Each control links to the official text, which is the authority. This is not legal advice.
Drafted by the Index Drafted automatically from the official text, not yet reviewed by a person.Reviewed Checked by a person against the official text.
Themes covered 0 of 35
Choose a theme to see only its controls. About the themes
Risk and scope 0 of 4
- Risk classification, not covered
- Prohibited practices, not covered
- Risk management system, not covered
- Impact assessment, not covered
Data and privacy 0 of 4
- Data governance and quality, not covered
- Data subject rights and consent, not covered
- Cross-border data transfer, not covered
- Copyright and training data, not covered
Transparency and records 0 of 4
- Transparency and disclosure to users, not covered
- Labelling of synthetic content, not covered
- Technical documentation, not covered
- Record keeping and logging, not covered
Technical safeguards 0 of 4
- Human oversight, not covered
- Accuracy, robustness and cybersecurity, not covered
- Bias and non-discrimination, not covered
- Testing and evaluation, not covered
Assurance and lifecycle 0 of 4
- Conformity assessment and certification, not covered
- Registration and notification, not covered
- Incident and breach reporting, not covered
- Post-market monitoring and corrective action, not covered
Rights and protected groups 0 of 4
- Automated decision-making rights, not covered
- Children and vulnerable groups, not covered
- Biometric identification and surveillance, not covered
- Whistleblowing, complaints and redress, not covered
Roles and organisation 0 of 6
- Supply chain and provider-deployer duties, not covered
- General purpose and foundation models, not covered
- Accountability and governance roles, not covered
- Training and AI literacy, not covered
- Public sector use and procurement, not covered
- Innovation support and sandboxes, not covered
Enforcement and sectors 0 of 5
- Enforcement, penalties and regulators, not covered
- Sector rules: health, not covered
- Sector rules: finance and insurance, not covered
- Sector rules: employment, not covered
- Sector rules: education, not covered
The full list of controls could not be loaded, so only the first 0 are shown.
Change history
No changes recorded since we began monitoring this policy.