Would Artificial Intelligence, Like ChatGPT, Be a Good ‘Peer’ Reviewer in Academic Publishing? A Human Versus AI-Based SWOT Assessment

Abstract

With the continuing passionate debate about the role of ChatGPT, an artificial intelligence (AI)–based text generator, in academic research and publishing, we reflect on whether AI can serve as a peer reviewer to overcome human weaknesses and other inherent weaknesses of the current human-supported peer-review model. The authors made their own assessment of this possibility by conducting a strength, weaknesses, opportunities, and threats (SWOT) analysis of human-based peer reviewers versus AI-based reviewers on 12 November 2023. A similar SWOT analysis was conducted for both humans and AI by employing cues fed to the paid version of ChatGPT (GPT-4). Despite the authors’ own ample experience and considerable efforts to make a balanced SWOT analysis, ChatGPT-4 was able to provide a comprehensive assessment, although it refused to cite relevant literature and obtained its information from websites and blogs. An earlier SWOT analysis (15 February 2023) using the free version (GPT-3) erred in citing the relevant literature while some citations and references were fabricated. Although AI (in this case, GPT-4) provided a logical and reasonable SWOT analysis, demonstrating its strength, it has not reached—in the authors’ view—a dependable stage yet to review articles for trusted academic journals.

Ähnliche Arbeiten