Expanding Digital Linguistic Representation
UCT Leads Project to Build AI for African Languages
Researchers focus on large language models for isiXhosa, isiZulu, and Sepedi to preserve digital heritage.

Researchers at a university desk looking at a monitor displaying indigenous South African text and AI data structures.
Photo: Avantgarde News
The University of Cape Town (UCT) is spearheading a national collaboration to develop artificial intelligence tools for African languages [1]. This multi-university effort focuses on creating large language models specifically for isiXhosa, isiZulu, and Sepedi [1]. The project aims to ensure these languages remain preserved and represented as digital technology evolves [1]. A broader study involving 7,000 Africans highlights the ongoing necessity for human involvement in AI development [2]. Researchers emphasize that local context is vital for creating effective digital tools that resonate with users [2]. This initiative seeks to bridge the gap in linguistic representation within the global AI landscape [1].
Editorial notes
Transparency note
Drafted with LLM; human-edited
- AI assisted
- Yes
- Human review
- Yes
- Last updated
Risk assessment
The risk level was elevated to high because the provided source list contained only two independent domains, falling below the recommended minimum of three.
Sources
Related stories
View allTopics
About the author
Avantgarde News Desk covers expanding digital linguistic representation and editorial analysis for Avantgarde News.


