How AI use in scholarly publishing threatens research integrity, lessens trust, and invites misinformation

Since 2023, a significant number of published scholarly papers show signs of having been edited using AI tools. Many of these will have been cases where it was used for copyediting or translation, but some have been substantially or entirely generated by large language models—and readers cannot easily tell which is which. AI tools, meanwhile, are also being used to review papers and as intermediaries in search and discovery tools, in ways that are not always well understood, and academic databases are starting to show signs of being affected by AI-generated hallucinated research. These tools are placing additional stresses on the integrity of the scholarly publishing system, potentially making it more vulnerable to organized and intentional disinformation campaigns.