New guidelines issued by the Ministry of Science and Technology prohibit researchers from using generative AI to directly generate declaration materials for their research or having AI be listed as a co-author of research results.  

Released on Dec. 21, the research code of conduct applies to researchers in scientific institutions, higher education institutions, medical institutions, and enterprises.

The ministry said the guidelines are a response to new challenges in research data processing and intellectual property rights that have arisen from the rapid development of AI.

The guidelines require all AI-generated content to be clearly labeled as such, with information provided as to how the content was generated. 

Zhang Xin, director of the Digital Economy and Legal Innovation Research Center at the University of International Business and Economics in Beijing, believes the guidelines will help promote more responsible use of generative AI in scientific research. 

“If researchers use reference materials generated by AIGC (AI-generated content) without verification, it may not only jeopardize the quality of research outcomes but also intensify the spread of false information, posing various risks to society,” Zhang told Sixth Tone.

The prohibiting of generative AI as a co-author aligns with broader academic practice in China currently, Zhang added.

In September, the Institute of Scientific and Technical Information of China, a research institute under the Ministry of Science and Technology, collaborated with world leading academic publishers Elsevier, Springer Nature, and John Wiley & Sons to release guidelines on the use of AI-generated content in academic papers, which also required clear labeling of such content. 

In August, authorities released an updated draft law on academic degrees specifying that students caught using AI to write dissertations will have their degrees revoked. While the draft has yet to be finalized, some domestic academic journals are already rejecting papers produced with the help of generative AI.  

Further clarification of rules surrounding the use of AI in research is needed as AI becomes an important research tool, said Zhang. 

As AI tools continue to proliferate in China, authorities have been proactive in regulating various applications, including recommendation algorithms and “deepfakes” — fake videos or recordings of people manipulated through AI. 

In April 2023, the Cyberspace Administration of China unveiled specific rules for generative AI, becoming the first in the world to do so.

Meanwhile, major social media platforms such as Douyin, the Chinese version of TikTok, and video streaming platform Bilibili have also started requiring labeling of AI-generated videos.

(Header image: IC)