Prevalence and Characteristics of Generative AI Policies in Vascular Surgery Journals: A Cross-Sectional Review

The integration of generative artificial intelligence (GenAI) into academic publishing presents new opportunities and ethical challenges. Transparent editorial oversight is essential, yet the extent of GenAI policy adoption among vascular surgery journals remains unclear. We conducted a cross-sectional review of vascular surgery journals indexed in the 2024 Scimago Journal Rank and Google Scholar Metrics. Eligible journals were active, peer-reviewed titles regularly publishing vascular surgery research. As of June 12, 2025, two independent reviewers assessed author instructions, journal websites, and publisher policies for GenAI-related guidance. Extracted variables included permitted GenAI applications (e.g., language editing, image generation), disclosure requirements, authorship restrictions, naming of specific tools, and human accountability clauses. Journal characteristics-geographic origin, SJR quartile, and H-index-were also recorded. Descriptive statistics and Welch's t-tests were used for analysis. Of 25 journals evaluated, 18 (72%) had explicit GenAI editorial policies. All prohibited attributing authorship to AI tools and emphasized human responsibility. Nineteen journals (76%) required disclosure of GenAI use, but only 14 (56%) specified disclosure location. Thirteen (52%) addressed AI-generated images. Journals permitting GenAI-assisted language editing had significantly higher H-indices (70.2 vs. 20.7; p = 0.0167). SJR Q1/Q2 journals demonstrated more comprehensive policies than lower-tier journals (6.8 vs. 4.0 elements; p = 0.0094). U.S. and U.K.-based journals led in policy specificity. While GenAI policies are increasingly present in vascular surgery journals, inconsistencies persist. Standardized, enforceable editorial guidance is needed to ensure ethical and transparent integration of GenAI into scholarly publishing.