Korin AI asks what music generation looks like when built in Nigeria

An open-access paper in AI & Society examines Korin AI, a Nigerian-owned music-generation platform built around African musical data, genres and languages. The authors draw on fieldwork in Lagos with musicians, producers, platform founders and entertainment lawyers.

Their argument is more useful than a standard product launch. Korin can lower production barriers and make musical traditions visible to models that often privilege Western categories. At the same time, it still depends on external cloud infrastructure, uneven access to data and decisions about which languages and forms receive support first.

“African music” is not one training category

The paper warns against treating representation as a box that can be checked by adding a generic African style. Nigeria’s music is heterogeneous, and technical systems can reproduce new exclusions even while correcting older ones.

Korin matters because it shifts the question from whether creators in the Global South will adopt music AI to who gets to build it, whose musical intelligence becomes legible to the system and which infrastructure controls the result. Those are product questions, but they are also cultural and economic ones.