Prevalence of generative artificial intelligence guidance statements in the urology literature: a descriptive study

Abstract

Purpose: The adoption of generative artificial intelligence (AI) in medical literature has increased exponentially over the past 2 years. Many journals have introduced AI guidance statements for authors during the manuscript submission process. This study characterizes the extent and types of AI guidance statements among urology journals.Methods: A total of 112 urology journals indexed on PubMed were identified. Each journal’s website was searched for the presence of an AI guidance statement. Specific aspects of AI guidance assessed included manuscript content generation, manuscript writing, and manuscript editing. Additional variables such as journal data, region, subspecialty, society affiliations, and impact factor were also collected.Results: Of the total 112 urology journals, 61 (54.5%) had an AI guidance statement. Most journals with statements (n=58, 95.1%) permitted the use of AI for manuscript editing. A slightly smaller majority (n=53, 86.9%) explicitly allowed AI-assisted manuscript writing. No journals definitively prohibited AI use for manuscript editing. Twenty-three journals (37.7%) permitted AI-generated manuscript content, while 11 (18.0%) explicitly did not, and 27 (44.3%) were unclear regarding their stance. Among journals with any AI usage, 60 (98.4%) required a disclosure statement on AI use. Only one journal (1.6%) did not provide any guidance.Conclusion: More than half of urology journals offer author guidance on the use of AI in manuscript submission. However, these instructions are not standardized across journals. As AI continues to permeate medical literature, the development of consensus policies is advisable.

Ähnliche Arbeiten