Research Integrity and Generative Artificial Intelligence: An Oxymoron?

Résumé

The prevalence of Generative AI (GenAI) poses several challenges. A study involving more than 1600 scientists found that almost 30% employed GenAI in writing papers, with another 15% in grant applications (Prillaman, 2024). The heavy influence of LLMs in academic writing is evident in an analysis of 950,965 papers between 2020 and 2024, with the swiftest and largest increase of 17.7% in Computer Science. Indeed, fraudulent research practices have increased since the launch of ChatGPT (Olojede, 2024). An overarching concern GenAI has brought into higher education is the issue of academic integrity. But what does research integrity specifically mean in the age of AI? Using philosophical tools of argument, critical thinking, and reconstruction of ideas, this paper argues that while the developmental process of GenAI is anathema to the principles of research integrity, the human using the tools can attempt to salvage the situation by adhering to integrity principles.