Handling multilingualism by the EU institutions at the time of AI-based language technologies

This article investigates how artificial intelligence (AI)-driven language technologies employed by European Union (EU) institutions affect the relationship between AI and the EU’s respect for linguistic diversity. It emphasises the necessity for better protection and representation of under-resourced languages, in particular minority or minoritised languages, in the face of rapid technological advancement. Systemic disadvantages may be faced by regional, minority, and migrant languages and even official EU languages despite their cultural significance. The study identifies key challenges of AIpowered language technologies, such as scarce data resources, tendencies towards linguistic homogenisation, unequal development of language technologies and reinforcement of cultural stereotypes. The findings reveal the pressing need for responsible AI practices within the EU institutions that promote inclusivity, targeted policies, community engagement, interdisciplinary collaboration, and strict human oversight to mitigate bias and ensure fairness. The analysis underlines the critical role of policy initiatives and active involvement of linguistic communities in shaping such AI tools that genuinely promote and uphold linguistic diversity. Only by embracing these principles, the EU can strive for cultivating more inclusive digital spaces and thereby preserve Europe’s rich multilingual legacy.