Cultural Globalization and the Legal Protection of Linguistic Diversity in the Age of Generative Artificial Intelligence

Abstract

Generative artificial intelligence is rapidly becoming a major infrastructure for communication, education, knowledge production, translation, journalism, professional services, and digital cultural participation. However, contemporary language models are characterized by substantial inequalities in the availability, quality, and representativeness of linguistic data. While globally dominant languages benefit from extensive digital resources, many minority, indigenous, regional, and low-resource languages remain underrepresented. This disparity raises questions extending beyond technological performance to cultural rights, linguistic equality, and the distribution of opportunities within globalization. This article investigates whether the emerging governance of generative artificial intelligence should incorporate positive obligations concerning linguistic diversity. Drawing on international human rights law, cultural rights, minority protections, intellectual property principles, and emerging artificial intelligence regulation, the study examines the legal implications of language-based algorithmic exclusion. It argues that unequal linguistic representation can limit access to public information, digital education, automated services, political participation, and cultural production, thereby reinforcing existing patterns of global linguistic hierarchy. At the same time, large-scale collection of minority-language materials for AI training may generate concerns regarding community consent, cultural appropriation, copyright, sacred knowledge, and exploitation of linguistic resources. The article therefore proposes a framework of linguistic justice for generative AI based on meaningful accessibility, dataset transparency, community participation, culturally appropriate data governance, support for low-resource language development, and special safeguards for restricted cultural knowledge. Rather than requiring identical technological performance across all languages, the framework imposes differentiated obligations according to the social importance and risk profile of the AI application concerned. The study demonstrates that linguistic diversity should not be treated merely as a question of technological localization or market expansion. It is increasingly a legal issue concerning equality, cultural participation, and access to the institutional infrastructures of contemporary society. Protecting linguistic pluralism should consequently become an integral component of global AI governance.