Regulation of use of generative AI in Diamond Open Access Journals: a comparative analysis of editorial policies

Abstract

The rapid expansion of generative artificial intelligence has transformed the processes of production, writing, and communication of scientific knowledge, raising new challenges related to authorship, academic integrity, and transparency in scholarly publishing. In this context, this study examines how Latin American diamond open access journals regulate the use of artificial intelligence tools within their editorial processes. A comparative documentary analysis of editorial policies was conducted on technology journals indexed in the Directory of Open Access Journals (DOAJ). The sample consisted of 50 Spanish-language diamond open access journals. For each journal, editorial policies, codes of ethics, author guidelines, and other regulatory documents available on their official websites were reviewed. The identified policies were analyzed using a structured coding matrix that included variables related to mandatory AI disclosure, authorship rules, peer review regulation, control mechanisms, sanctions, and alignment with international standards. The results show that only 20% of the analyzed journals explicitly mention the use of generative artificial intelligence. Among these journals, 90% require explicit disclosure of AI use and 90% include editorial control mechanisms, primarily similarity detection tools. However, only 40% have independent structured policies and the regulation of AI use by different editorial actors appears only partially. The analysis identified three regulatory models: basic declarative (40%), intermediate (40%), and advanced (20%). The findings reveal conceptual convergence with international standards regarding transparency and human responsibility, but also highlight a gap in the formalization of regulatory frameworks within the Latin American diamond open access ecosystem.