In a recent Nature article, James Zou (Stanford University) calls for guidelines on using ChatGPT and other large language models (LLMs) in peer review. The author notes that since ChatGPT’s 2022 release, computer scientists have observed an increase in the number of chatbot-generated reviews, which often lack depth and overlook errors. While Zou sees potential for LLMs in certain review tasks—such as retrieving related papers or turning vague suggestions into actionable edits—he argues that clear guidelines are essential. Zou urges journals to require reviewers to disclose their LLM use and to encourage more human interactions throughout the peer review process. “Establishing community norms and resources will help to ensure that LLMs benefit reviewers, editors and authors without compromising the integrity of the scientific process,” concludes Zou.
