From postponed tour to platform: Nkenne’s Zoom-fueled mission to preserve African languages
Nobody is talking about this: 2,000+ African languages are disappearing from AI training data while the industry chases English.
Why it matters
As AI labs race to scale models, linguistic diversity gaps are widening. Nkenne's platform addresses a systemic blind spot in AI development that affects billions of people and reveals how commercial incentives are reshaping which languages survive in the digital age.
The key facts
10 to knowAfrican languages historically underrepresented in AI training datasets
Platform built to preserve tonal, hyper-local, deeply contextual languages
Founded as response to AI's English-default architecture
Leverages Zoom infrastructure for distributed language documentation
Highlights gap between AI capability (multilingual claims) and reality (language coverage skew)
African languages underrepresented in major AI training datasets
Thousands of tonal, hyper-local African languages lack AI/ML infrastructure
Founded on mission to preserve and digitize African languages via platform approach
Using Zoom-based collection methodology for distributed language data gathering
Addresses AI equity gap in language model development
Go to the source
SiliconAnglesiliconangle.com
Publisher excerpt: When people talk about artificial intelligence and language, the focus usually defaults to English, a handful of European languages and perhaps a few from Asia. African languages – thousands of them, often tonal, hyper-local and deeply contextual – rarely make the roadmap. That blind spot is…