China’s TC260 Releases Ethics-Safety Guidelines for Artificial Intelligence Applications 1.0
Published 21 May 2026
Yu Du
On 19 May 2026, the National Technical Committee 260 on Cybersecurity of Standardization Administration of China (TC260) released Ethics-Safety Guidelines for Artificial Intelligence Applications 1.0 (Guidelines). In China’s standardization system, “TC” stands for “Technical Committee”, while “260” is the committee’s official numerical designation under the Standardization Administration of China (SAC). TC260 is one of China’s most influential standard-setting bodies in the fields of cybersecurity, data governance, personal information protection, and artificial intelligence governance.
The document was issued by TC260 against the backdrop of the rapid development of generative artificial intelligence, AI agents, and other emerging AI technologies. TC260 noted that while AI is reshaping society, production, and daily life with unprecedented efficiency and convenience, it is also creating significant ethical and safety challenges. The Guidelines aim to promote “safe, controllable, and beneficial” AI development by balancing innovation, security, and ethical considerations across different application scenarios.
The Guidelines are described as a principled and reference-oriented technical document. They are intended to provide guidance for organizations and individuals engaging in AI-related activities, while also serving as a reference for regulators, industry associations, and governance institutions.
Artificial Intelligence Ethics-Safety Impacts
The Guidelines identify several major ethical and safety risks arising from AI applications. According to TC260, AI systems may challenge human control and governance if they operate beyond the scope of effective human supervision. The document also highlights risks relating to public order, social trust, cognitive dependence, discrimination, and unequal treatment of different social groups. In addition, the Guidelines emphasize that AI technologies may affect fundamental rights and interests, including privacy, dignity, health, property, and personal safety, and may even create broader ecological and sustainability concerns if deployed irresponsibly.
Ethics-Safety Concepts and Principles
To address these risks, the Guidelines establish a comprehensive framework of ethical and safety principles. The document stresses that AI development should remain people-centered and oriented toward improving human well-being and social welfare. It also emphasizes respect for life, human dignity, fairness, transparency, privacy protection, and meaningful human oversight over AI systems. At the same time, the Guidelines advocate balancing innovation and security, strengthening risk control, promoting international cooperation, and encouraging inclusive and open-source AI ecosystems that allow broader sharing of technological benefits.
Ethics-Safety Guidance
The Guidelines further provide practical guidance for AI developers, service providers, and users. Developers are encouraged to integrate ethical and safety considerations throughout the entire AI lifecycle, including data selection, algorithm design, testing, deployment, and maintenance, while paying special attention to hallucinations, transparency, and protection of vulnerable groups. Service providers are advised to maintain human oversight in high-risk scenarios, establish mechanisms allowing users to reject or terminate AI services, minimize unnecessary collection of personal information, and strengthen emergency response capabilities. Meanwhile, users are encouraged to improve AI literacy, avoid overreliance on AI systems, use AI lawfully and responsibly, and carefully protect sensitive personal information and trade secrets when interacting with AI applications.
Comment
The release of the Guidelines represents another important step in China’s evolving AI governance framework. Although the document is framed as a reference-oriented technical guideline rather than a binding regulation, it provides a comprehensive statement of policy expectations regarding the ethical and safe deployment of AI technologies. The Guidelines place strong emphasis on “human control”, public order, and collective social interests, reflecting China’s broader governance approach toward emerging technologies. The explicit endorsement of open-source ecosystems and collaborative governance suggests that China is seeking to balance innovation promotion with regulatory oversight.
For developers, service providers, and multinational companies operating in China, the Guidelines may serve as an important indicator of future regulatory expectations and enforcement trends. They are also likely to influence industry standards, compliance practices, and future sector-specific AI governance measures in China.
The document was issued by TC260 against the backdrop of the rapid development of generative artificial intelligence, AI agents, and other emerging AI technologies. TC260 noted that while AI is reshaping society, production, and daily life with unprecedented efficiency and convenience, it is also creating significant ethical and safety challenges. The Guidelines aim to promote “safe, controllable, and beneficial” AI development by balancing innovation, security, and ethical considerations across different application scenarios.
The Guidelines are described as a principled and reference-oriented technical document. They are intended to provide guidance for organizations and individuals engaging in AI-related activities, while also serving as a reference for regulators, industry associations, and governance institutions.
Artificial Intelligence Ethics-Safety Impacts
The Guidelines identify several major ethical and safety risks arising from AI applications. According to TC260, AI systems may challenge human control and governance if they operate beyond the scope of effective human supervision. The document also highlights risks relating to public order, social trust, cognitive dependence, discrimination, and unequal treatment of different social groups. In addition, the Guidelines emphasize that AI technologies may affect fundamental rights and interests, including privacy, dignity, health, property, and personal safety, and may even create broader ecological and sustainability concerns if deployed irresponsibly.
Ethics-Safety Concepts and Principles
To address these risks, the Guidelines establish a comprehensive framework of ethical and safety principles. The document stresses that AI development should remain people-centered and oriented toward improving human well-being and social welfare. It also emphasizes respect for life, human dignity, fairness, transparency, privacy protection, and meaningful human oversight over AI systems. At the same time, the Guidelines advocate balancing innovation and security, strengthening risk control, promoting international cooperation, and encouraging inclusive and open-source AI ecosystems that allow broader sharing of technological benefits.
Ethics-Safety Guidance
The Guidelines further provide practical guidance for AI developers, service providers, and users. Developers are encouraged to integrate ethical and safety considerations throughout the entire AI lifecycle, including data selection, algorithm design, testing, deployment, and maintenance, while paying special attention to hallucinations, transparency, and protection of vulnerable groups. Service providers are advised to maintain human oversight in high-risk scenarios, establish mechanisms allowing users to reject or terminate AI services, minimize unnecessary collection of personal information, and strengthen emergency response capabilities. Meanwhile, users are encouraged to improve AI literacy, avoid overreliance on AI systems, use AI lawfully and responsibly, and carefully protect sensitive personal information and trade secrets when interacting with AI applications.
Comment
The release of the Guidelines represents another important step in China’s evolving AI governance framework. Although the document is framed as a reference-oriented technical guideline rather than a binding regulation, it provides a comprehensive statement of policy expectations regarding the ethical and safe deployment of AI technologies. The Guidelines place strong emphasis on “human control”, public order, and collective social interests, reflecting China’s broader governance approach toward emerging technologies. The explicit endorsement of open-source ecosystems and collaborative governance suggests that China is seeking to balance innovation promotion with regulatory oversight.
For developers, service providers, and multinational companies operating in China, the Guidelines may serve as an important indicator of future regulatory expectations and enforcement trends. They are also likely to influence industry standards, compliance practices, and future sector-specific AI governance measures in China.