The AI ‘Ghosts’ Contaminating Academic Publishing
“The academic record is being quietly haunted” by researchers with names like Elena Vasquez and Marcus Chen.
A new research paper by Samsung and the University of Warsaw has identified a troubling issue in academic publishing - "ghost authors" created by AI language models. These AI-generated names repeatedly appear in academic papers, articles, and books, even though the individuals behind the names never existed. The study, titled "The Ghost Couple: Correlated LLM Name Priors and Their Haunting of the Web and Academic Publishing," reveals that certain language models consistently generate the same names for experts in specific fields.
For example, the name Marcus Chen often emerges when users ask AI to create a software developer. The paper found that these "ghost" authors have co-authored hundreds of AI-generated documents, contaminating the web and academic publishing. The researchers discovered that some names are correlated, meaning certain combinations of names frequently appear together.
Names like Elena Amara Okafor, Aris Thorne, Lena Petrova, and Elara Voss have also been consistently generated by AI models like Claude, Gemini, and ChatGPT. These AI-generated names have been found in academic databases, fake journals, and research platforms like ResearchGate and Google Scholar. This proliferation of AI-generated content is causing contamination in the scholarly record, as these ghost authors appear on academic aggregators and are indexed by search engines without verification.
While the research offers a potential solution for detecting AI-generated content, the sheer volume of AI-generated material on the internet is likely to dilute the effectiveness of these methods over time.
Written by urgent.news from 404 Media's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
