The unfortunate large number of out-of-school children in Nigeria has rightly received wide attention and governments and other non-profit organizations are working towards reducing such menace. But another phenomenon that gets comparatively low coverage is that the majority of school goers are not learning, especially at elementary level. Research has shown that the likely causes include language barrier, the inability of educators to tailor teaching methods to fit the individual rather than the peer group, the lack of additional support outside the school environment, and the inability of government and school owners to adequately deploy supportive technology to aid learning. 

The aim of this research, therefore, is to develop a responsible, ethical and locally developed artificial intelligence (AI)-driven solution to influence and facilitate the processes of teaching and learning non-linguistic subjects in Nigerian primary schools, with mathematics as the case study. We aim to investigate best practices to develop and implement an interactive digital curriculum, including pictorial and audio-visual gamification techniques, in both English and the target local languages to enable pupils in primary school to learn basic arithmetic in the languages of their immediate environments. 

The solution will also enable students to learn at their own pace through adaptive AI and personalised learning techniques. In addition to the final solution, an AI Learning Assistant, we aim to release the digital content, including the curriculum, machine translation, text-to-speech and speech-to-text models and AI chatbot to enable further research.