Case Report of an Emerging Issue in Journal Publishing: Artificial Intelligence and the Critical Letter to the Editor

Abstract

<h3>Context</h3> Letters to the Editor (LttE) of journals are common means to advance scholarly conversations, and most biomedical journals include Letters as publications that are indexed in Medline or similar venues. With the advent of Artificial Intelligence and its ability to rapidly summarize content and write “human-sounding” manuscripts, it may be possible for authors to generate many LttEs by systematically combing the literature for new papers, having AI critique the text, and generate letters to the editor. At the PRiMER journal, we have encountered such an instance, and believe it important to report to the field, in advance of what may be an emerging practice. <h3>Objective</h3> To describe a possible instance instance of LttE manufacturing, and the methods by which we identified the phenomenon. <h3>Study Design and Analysis</h3> Case Report. <h3>Setting</h3> Family Medicine (STFM-Sponsored) medical journal, PRiMER. <h3>Intervention</h3> A LttE manuscript was assessed for content, found to be critical and incisive beyond what is normally observed post-review and publication. We examined the authors’ publication track record for area expertise, and used three online AI detectors to assess whether the letter had been produced by AI tools. <h3>Outcome Measures</h3> Qualitative assessment of author bibliographies on ORCID; AI content percentage. <h3>Results</h3> The authors of the LttE were found to be regularly and frequently publishing critical LttEs across multiple fields, disciplines, and methods employed. The AI content of the submitted LttE was determined to be between 46%-100% AI-generated. <h3>Conclusions</h3> In order to generate large numbers of publications to add to CVs, some authors appear to be generating both critical assessments, and subsequent LttEs, using AI tools. While legitimate discourse and criticism of published literature is vital to scientific discourse, the ability to rapidly generate critical LttEs for the sole purpose of filling CVs, without human scholarly expertise, may be a challenge for journal editors and disciplines as a whole.

Ähnliche Arbeiten