On AI and the future of African lexicography

Abstract

Artificial intelligence (AI) is increasingly influencing the development of lexicography and offers significant potential for transforming dictionary production and use in African languages. In multilingual contexts such as South Africa, where twelve official languages coexist within the education system, AI technologies present new opportunities for the creation of accessible, dynamic, and user-oriented lexicographic resources. This article examines the opportunities, capabilities, and challenges associated with the integration of AI in African lexicography, with particular attention to its potential role in supporting multilingual education. AI technologies, including machine learning and natural language processing, enable faster corpus analysis, automated identification of lexical patterns, and the efficient updating of lexical databases. These capabilities can support the development of lexicographic tools that move beyond static reference works to become interactive digital tools. Such dictionaries can incorporate features such as intelligent search, pronunciation support, contextual usage examples, and multilingual linking between African languages and English. In multilingual educational environments, these tools can assist learners in accessing disciplinary vocabulary across languages, thereby supporting reading comprehension and facilitating multidisciplinary learning. However, the application of AI in African lexicography is constrained by several structural challenges. A major limitation is the scarcity of large, balanced, and machine-readable corpora for many African languages. Much linguistic data remains undocumented, fragmented, or unavailable in digital form, particularly in relation to spoken language and emerging terminology. These data gaps can limit the accuracy and representativeness of AI-driven lexicographic systems. Addressing these challenges requires sustained efforts in corpus development, digitisation of existing resources, and the creation of accessible linguistic datasets. The article argues that while AI cannot replace the expertise and interpretive role of lexicographers, it offers powerful tools that can enhance lexicographic practice and expand access to dictionaries. When combined with ongoing linguistic research and data development, AI has the potential to support the creation of innovative, user-centred lexicographic resources that strengthen multilingual education, improve reading development, and promote the use and status of African languages.