HiNews
Researchers have unveiled a critical flaw in the popular AI tool, ChatGPT. The study, titled “Diagnosis Unreliability of ChatGPT for Journal Evaluation,” reveals that ChatGPT, despite its widespread use and acclaim, falls short when it comes to evaluating the quality of academic journals.
The Study’s Findings
Conducted by a team of international researchers, the study meticulously analyzed ChatGPT’s ability to distinguish between legitimate, predatory, and hijacked journals. The results were startling: ChatGPT was found to be unreliable in its assessments, often failing to accurately identify the quality of journals. This revelation has significant implications for the academic community, which increasingly relies on AI tools for research and publication decisions.
What is a hijacked journal?
A hijacked journal is a type of predatory publication where a malicious third party creates a counterfeit website that mimics a legitimate academic journal. The hijackers use the legitimate journal’s title, International Standard Serial Number (ISSN), and other metadata to deceive researchers into submitting their work to the fake journal.
These fraudulent websites often charge authors high fees for publication without providing proper peer review or editorial services. The goal is to exploit the open-access model for profit, damaging the reputation of the legitimate journal and misleading researchers.
What is a predatory journal?
Predatory Journals are completely fake journals that often promise rapid publication without proper peer review or editorial services. They exist solely to collect publication fees from authors.
Predatory and hijacked journals are a growing concern in the academic world. These journals exploit researchers by charging high fees without providing proper peer review or editorial services, leading to the dissemination of low-quality or even fraudulent research. The study highlights the urgent need for more reliable tools to combat this issue.
The Alternative Solution
In response to ChatGPT’s shortcomings, the researchers have developed an early trial version of a new tool called the Journal Checker Chatbot. This alternative aims to provide more accurate evaluations of journal quality, helping researchers avoid the pitfalls of predatory and hijacked journals.
The chatbot is accessible online via https://t.me/hijackedjournalcheckerbot
Implications for the Future
The findings of this study underscore the limitations of current AI technology and the need for continued innovation in the field. As AI tools become more integrated into academic workflows, ensuring their reliability and accuracy is paramount. The development of the Journal Checker Chatbot represents a promising step forward, but the journey to fully trustworthy AI in academia is far from over.
This study serves as a wake-up call for the academic community. While AI tools like ChatGPT offer incredible potential, their limitations must be acknowledged and addressed. The development of more reliable alternatives is crucial to maintaining the integrity of academic research and publication.
Reference:
Dadkhah M, Oermann MH, Hegedüs M, Raman R, Dávid LD. Diagnosis Unreliability of ChatGPT for Journal Evaluation. Adv Pharm Bull. 2024 Mar;14(1):1-4. doi: 10.34172/apb.2024.020. Epub 2023 Oct 14. PMID: 38585462; PMCID: PMC10997925.
