Systematic Plagiarism, Author Identity Misuse, and Fabricated Literature Citations in AI-Generated Herpetological Resources
Open Access
- 13 August 2026
- journal article
- Published by Society for the Study of Amphibians and Reptiles in Journal of Herpetology
- Vol. 60 (3) , 255-259
Abstract
The increasing use of artificial intelligence (AI) in scientific writing has introduced an underrecognized threat to the integrity of the scholarly literature through the production of invalid reference citations. This analysis presents a comprehensive audit of herpetological literature citations in the online encyclopedia Animal-pedia, a large resource focused on reptiles. Across approximately 1.4 million words of text, I identified extensive citation errors, including (1) plagiarism of published herpetological works through the removal and replacement of their authors' names; (2) misuse of scientists' identities through the attribution of titles to individuals who did not produce them; and (3) false representation of publication venues, including both real and nonexistent journals. In total, I document >450 invalid literature citations, involving >1,360 cases of plagiarized and false authorship attribution and >580 false assignments of publication sources. Misassigned publication venues included major herpetological journals; the Journal of Herpetology was the most frequently misused. Most errors involved the recombination of real bibliographic elements into false citations, producing references that appeared credible but were invalid. Many fabricated titles were accompanied in search results by detailed AI-generated summaries of the nonexistent papers, reinforcing their apparent legitimacy. These findings demonstrate that AI-generated citation errors can occur on a massive scale in widely accessible resources. Such invalid citations pose a serious threat to the reliability of the herpetological literature because of their ready exposure to AI systems and their potential to propagate through legitimate publications. Rigorous verification practices are essential to prevent the propagation of AI-generated invalid citations into the herpetological literature.Keywords
This publication has 24 references indexed in Scilit:
- Hallucinated citations are polluting the scientific literature. What can be done?Nature, 2026
- Can researchers stop AI making up citations?Nature, 2025
- Tackle fake citations generated by AINature, 2025
- The citation catastrophe: Propagation of AI-generated counterfeit citations in scholarshipThe Journal of Academic Librarianship, 2025
- AI hallucinations can’t be stopped — but these techniques can limit their damageNature, 2025
- Chain retraction: how to stop bad science propagating through the literatureNature, 2024
- Beware of references when using ChatGPT as a source of information to write scientific articlesAmerican Journal of Obstetrics and Gynecology, 2023
- The exciting potential for ChatGPT in obstetrics and gynecologyAmerican Journal of Obstetrics and Gynecology, 2023
- Recognition, reward and responsibility: Why the authorship of scientific papers mattersMaturitas, 2009
- Fostering integrity in research: Definitions, current knowledge, and future directionsScience and Engineering Ethics, 2006