Artificial intelligence (AI) is increasingly changing the landscape of scientific publishing, but concerns are being raised about the trustworthiness of the content. The use of AI-generated text and images in academic papers has become more prevalent, with examples including an infographic of a rat with a large penis and an introduction that starts with 'Certainly, here is a possible introduction for your topic'. While AI programs like ChatGPT can be helpful tools if thoroughly checked and disclosed, there have been cases where AI-generated content slipped past peer review [c8a27b81].
The presence of AI-generated content in scientific journals has led to the proliferation of what is being referred to as 'junk' papers. These papers include poor quality, plagiarized, and fake content produced by paper mills. The demand for more papers puts pressure on academics, creating a 'vicious cycle' that pushes researchers to publish more. This raises concerns about the integrity and reliability of scientific research [c8a27b81].
In response to these challenges, China has proposed new guidelines requiring conspicuous labels for AI-generated content across various media, including text, video, audio, and virtual scenes. This initiative aims to combat misinformation, deepfakes, and copyright issues, reflecting a global trend towards regulating AI-generated content. The draft regulations encourage the use of subtle identifiers like digital watermarks and prohibit tampering with labels [20d3d149].
While AI translation tools can be valuable for non-native English speakers, there are concerns about errors, inventions, and unintentional plagiarism by AI. These issues could erode trust in science and the credibility of research findings. It is crucial for researchers, publishers, and peer reviewers to thoroughly check and disclose the use of AI-generated content in scientific publishing to maintain the integrity of the field [c8a27b81]. The public consultation on China's proposed regulations is open for one month, building on previous regulations from January and August 2023, and aims to enhance governance and reduce misinformation [20d3d149].