|
Getting your Trinity Audio player ready...
|
The artificial intelligence (AI) revolution is here, as it is currently transforming everything from finance to healthcare. Yet, for Africa, the most profound impact may be on its oldest asset: the African literary canon – the vast repository of written classics, diverse indigenous languages, and vibrant oral traditions that form the continent’s intellectual bedrock.
As such, the central question for policymakers and cultural custodians today is whether AI will act as a digital preserver or a potent agent of cultural erasure. This is not merely a philosophical debate; it is a critical matter of economic and cultural sovereignty in the digital age.
The immediate threat of erasure stems from data scarcity. Global AI models are trained overwhelmingly on text and information that is predominantly English and Western-centric.
This creates a systemic bias that renders African heritage digitally fragile. Of the African continent’s estimated 2,000 languages, only a handful are deemed “high-resource” by major tech platforms, leaving the rest to languish in obscurity. When indigenous languages are absent from the AI’s training data, the technology cannot effectively process, transcribe, or generate content in those tongues, which accelerates linguistic decay.
Read also: Can Artificial Intelligence ever truly write Literature?
The silence of AI on African terms is a form of algorithmic marginalisation. It poses a direct threat to oral traditions, which are the very mechanisms through which knowledge, history, and moral codes are passed down. The stories, proverbs, and epic poetry of countless cultures remain trapped in audio formats, which are invisible to the AI tools that govern modern information retrieval. If the next generation relies on AI for education and information, and it is unable to recognise the voices of their ancestors, that cultural heritage is effectively lost. Furthermore, relying on foreign-trained AI to interpret African classics – the works of Soyinka, Achebe, and Aidoo – risks a form of digital colonialism. AI, devoid of deep cultural context, may flatten and standardise these complex narratives, stripping the soul of the text and replacing it with a diluted, algorithmically palatable imitation.
Despite the existential risks, AI offers the most powerful tools in history for cultural preservation and revitalisation. This technology can convert fragile, intangible heritage into durable, accessible, digital assets – a move that has immense educational and economic value.
For the oral traditions, AI-driven Automatic Speech Recognition (ASR) is a game-changer. It allows for the mass transcription, indexing, and translation of vast archives of spoken history, effectively creating immortal digital libraries. As such, an elder’s history, once limited to local memory, can be transformed into a searchable, cross-referenced global asset, safeguarding wisdom against the inevitable passing of generations.
Crucially, AI holds the key to the revitalisation of indigenous languages. African-led initiatives, such as the Masakhane movement, are actively building Natural Language Processing (NLP) models for these tongues. These models can power language learning apps, intelligent translation services, and educational resources, making it easier for younger generations to reconnect with their roots. By enabling education and commerce to be conducted in local languages, AI can transform them from subjects of academic study into drivers of a thriving digital economy.
Finally, for African classics, AI enhances accessibility and global scholarship. The technology can facilitate rapid translation, enabling works to reach new academic audiences and book markets worldwide. Similarly, it can perform complex textual analysis, which enriches our appreciation of the canon’s intellectual sophistication.
The choice between preservation and erasure is not a technical inevitability. It is a policy decision. For AI to serve the African canon, there must be a deliberate, concerted push for African agency in the AI ecosystem. This requires that we treat African linguistic and cultural data as the new oil of the 21st century. Governments and private capital must fund mass digitisation projects and the ethical collection of high-quality data in indigenous languages, whilst ensuring they are not just consumers, but producers of this vital commodity.
We must invest in the digital infrastructure and human capital necessary to train and deploy African-centric AI models. Furthermore, it is paramount to develop regional regulatory and ethical guidelines that ensure data ownership and cultural representation are non-negotiable features of any AI deployed on the continent. To passively wait for foreign-trained AI to accommodate Africa is to consent to the eventual marginalisation of the continent’s intellectual history. We must harness this technological wave not just to participate in the global economy, but to actively build an AI tool that speaks, remembers, and cherishes the African canon. This buttresses the fact that the fate of our literary heritage – our collective memory – is not in the code, but in our will to code our own future.
Akinrinde is a literary critic and fictional writer.
