Rethinking Peer Review in the Age of Artificial Intelligence: A Tiered Model for Health Promotion Journals

The number of papers published in health promotion journals has grown substantially for two decades, and mainstream use of generative artificial intelligence (AI) has accelerated that growth. Yet peer review remains fundamentally unchanged. Review still depends on volunteer time, and many journals prohibit reviewers from using AI to protect confidential, unpublished work. The result is an asymmetry: authors write faster with AI, while reviewers face mounting demand without the same tools. Funding agencies have begun responding; the NIH, for example, now caps annual application submissions per principal investigator. Journals cannot impose comparable limits because submissions are scattered across thousands of independent outlets, and requiring authors to complete reviews in proportion to their publications would be nearly impossible to enforce. We propose a solution that concentrates limited reviewer capacity on the most complex and consequential research. Cross-sectional and other descriptive studies would be evaluated by a purpose-built AI reviewer, trained to assess methodological rigor in such studies and designed not to retain or learn from unpublished manuscripts. Accepted papers would appear in a companion journal or section of the parent title. Traditional peer review would then be reserved for longitudinal, experimental, and other higher-impact research. Beyond easing reviewer burden, this model creates a transparent signal of rigor that could help tenure and promotion committees reward quality over publication quantity.