Artificial Intelligence (AI) is increasingly being incorporated into academic publishing to enhance the peer-review process. This includes automating the initial screening of manuscripts, assessing methodological rigor, detecting plagiarism, evaluating statistical accuracy, and even predicting the potential impact of submitted articles. However, the effectiveness and reliability of AI-based tools compared to traditional human-led review processes remain topics of active investigation. This question aims to explore existing scientific evidence supporting the efficacy, accuracy, and practical utility of AI assistance in peer-review, as well as to identify documented benefits, limitations, or gaps in current knowledge
As of now, most journals look down upon reviews done by Gen AI. However, the processes you have mentioned may be automated some day using AI. At least the basic screening.
Arindra Nath Mishra I agree. I have heard reviewers discuss this topic, each with their own perspectives and nuances. In fact, what is currently lacking is concrete evidence "on the table," increasingly supported by research, ultimately leading to systematic reviews on the subject.
As for the essence of research, artificial intelligence does not have its own life experience to prepare judgments. I think we should always consider the source of his opinion. To analyze articles, artificial intelligence can collect and compare information.
Scientific evidence shows that AI can assist in the peer-review process by automating tasks such as manuscript screening, assessing statistical accuracy, detecting plagiarism, and predicting the potential impact of articles. These tools can enhance efficiency, allowing reviewers to focus on more complex aspects. However, AI cannot replace human judgment, especially in subjective evaluations and understanding context. AI is most effective as a complement to human reviewers, not a complete replacement. Further research is needed to fully refine AI’s role in academic publishing.
- Badges
- Science topic