As generative AI becomes deeply integrated, the academic community is rapidly tightening the boundaries of "AI assistance."

Recently, multiple universities, including University of Science and Technology of China, Jiangsu University Medical School, and Hebei Institute of Environmental Engineering, have issued notices to formally regulate the use of AI tools in degree papers by graduate and undergraduate students. This marks a shift in the review of originality of academic papers from simple "plagiarism checks" to "precise compliance" with AI-generated content.

Clear Red Lines: Core Ideas Must Be "Original from Human Brains"

Although major universities have not completely banned AI, they have set strict restrictions on the "soul" parts of the papers. According to a notice released by the Graduate School of University of Science and Technology of China, the use of AI tools must follow the "human-led, AI-assisted" principle. The following content is explicitly prohibited from being generated by AI:

Core Elements: Core ideas and research conclusions of the paper.

Technical Approach: Research methods and design of research plans.

Original Data: It is strictly forbidden to use AI to generate or alter original data or original result images.

Graded Management: Which Sections Allow "AI Assistance"?

While tightening the management of core sections, some universities have also provided reasonable space for AI assistance. The Academic Affairs Office of Hebei Institute of Environmental Engineering mentioned in its document that students can use AI within the following scope in compliance:

Preparation Phase: Literature search and data collection.

Visual Aid: Assistance in creating charts related to the paper.

Language Polishing: Assisting with text formatting and grammar correction (subject to specific regulations of each university).

New Normal in Academic Integrity: Transparency and Controllability

The Medical School of Jiangsu University emphasized in its notice that using AI tools should follow the "auxiliary, transparent, and controllable" principle. This means that in the future, students may need to actively declare the scope of AI usage when submitting their papers to ensure the authenticity and reliability of the research process.