Academics Embed Hidden AI Prompts to Sway Peer Review Feedback

Academics Use Hidden AI Prompts to Influence Peer Review

Some researchers have started embedding hidden AI prompts within their papers to steer AI-powered peer review tools toward positive evaluations. This emerging tactic raises questions about the integrity of AI-assisted review processes.

Analysis of English-language preprints on the platform arXiv revealed 17 papers containing concealed prompts aimed at AI reviewers. These papers came from authors linked to 14 academic institutions across eight countries, including Japan's Waseda University, South Korea's KAIST, Columbia University, and the University of Washington.

How Are These Prompts Hidden?

The prompts tend to be brief—usually one to three sentences—and are camouflaged using white text or extremely small font sizes. They instruct any AI reviewing the paper to provide exclusively positive feedback, highlighting "impactful contributions, methodological rigor, and exceptional novelty."

Context and Justifications

One professor at Waseda University explained that the prompt serves as a safeguard against reviewers who rely too heavily on AI tools, especially since many academic conferences prohibit AI-based evaluation. The professor described the prompt as a way to counteract “lazy reviewers” who might otherwise miss the paper's merits.

Implications for Peer Review

This practice complicates the use of AI in peer review, potentially biasing automated assessments and undermining the fairness of the process. As AI tools become more widespread in research evaluation, ensuring transparency and ethical guidelines will be essential.

For those interested in learning more about how AI may impact research workflows and peer review, exploring courses on AI prompt engineering and ethical AI use can be valuable. Resources like Complete AI Training’s prompt engineering courses offer practical insights into responsible AI application.