Answers are grounded in this paper and cite their sources.
Experimental evaluation assesses AI text identification reliability in scientific literature, suggesting improvements are needed.
This is an English-language abstract. The article is devoted to evaluating the effectiveness of existing systems for identifying AI-generated text in scientific papers. The study analyzed six systems for their ability to detect AI-generated text in academic works. The probability scores for AI generation varied from 5 to 67% for the same text, depending on the system, method, and conditions of the check. The key factors leading to false results were the stylistic features of academic writing. The research calls into question the objectivity of current tools and emphasizes the need to develop specialized methods adapted to the scientific style of writing.
No takes yet. Share an insight, caveat, or question.
Savelev et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: