Guideline calls for human-centric AI

China has released a technical guideline on the ethics and safety of artificial intelligence applications, offering a Chinese approach that emphasizes human-centered development, AI for good, and human leadership over AI systems.

The document, titled "Ethics-Safety Guidelines for Artificial Intelligence Applications 1.0", was released on Tuesday at a subforum of the 2026 China Internet Civilization Conference in Nanning, capital of the Guangxi Zhuang autonomous region.

Issued by the National Technical Committee 260 on Cybersecurity, China's key body responsible for cybersecurity standardization, the guideline is described as a "principle-based and reference technical document". It covers AI application development, service provision and use, and provides guidance for organizations and individuals carrying out AI-related activities.

The document was drafted by leading Chinese institutions and companies, including Tsinghua University, the China Electronics Standardization Institute, technology giants Alibaba Group and Huawei Technologies, and DeepSeek, a well-known AI enterprise.

It outlines major principles including reasonable risk control, openness and transparency, privacy and security protection, controllability and trustworthiness, agile co-governance, and inclusive sharing.

Fan Kefeng, vice-president of the China Electronics Standardization Institute, said AI applications are expanding from content generation to assisted decision-making, interactive control and other fields. With the emergence of new forms such as intelligent agents and integrated AI systems, AI applications are becoming increasingly autonomous, complex and influential, he said.

"While improving production efficiency, AI has also made ethical and safety issues such as job displacement, decision-making responsibility and emotional dependence more prominent," Fan said, adding that such changes are bringing new challenges to human agency, social trust and public order.

The guideline states that AI applications should always serve "the common well-being of humanity", calling for ensuring that "the leadership over AI applications belongs to humans", with mechanisms for human control, emergency response and intervention established at key stages.

The document also urges relevant parties to retain "human judgment, supervision, intervention and correction" in key processes to avoid excessive replacement of human decision-making by AI.

For users, it advises moderate AI use and a proper understanding of AI emotional services. It states that AI should be treated as "a tool to assist real life", urging users to avoid excessive dependence, addiction, or replacing real interpersonal communication and activities with AI.

For information and communication services, the guideline calls for stronger risk governance of AI-generated and recommended content, warning against "information cocoons, cognitive misleading and cognitive degradation".

The guideline also encourages building an open-source innovation ecosystem, while improving the security capabilities of the open-source ecosystem.

Zhang Linghan, director of the institute of AI law and governance at China University of Political Science and Law, said the guideline provides detailed pathways and a Chinese approach to preventing ethical and safety risks at a time when AI is increasingly embedded in production and daily life, reshaping human-machine relations and being applied in more complex scenarios.

She said the guideline emphasizes that human leadership must not be weakened.

Beyond interpersonal and sociopolitical relations, it also addresses ecological concerns. "It also takes into account AI's consumption of energy and resources," said Zhang, who is also an expert with the United Nations High-Level Advisory Body on Artificial Intelligence.

Zhou Jingren, a partner at Alibaba Group and senior AI executive, said AI technology and large language models are still undergoing rapid development. From the outset, Alibaba established an ethics committee for large language models and invited experts from different fields to jointly explore solutions in this emerging area, he said.

Cheng Zixin contributed to this story.

Contact the writers at jiangchenglong@chinadaily.com.cn