As technology and AI become increasingly woven into the fabric of our daily lives, we must ask ourselves some fundamental questions: whose voices are being heard, and whose stories are being told, especially on the African continent?
May 25th is Africa Day, a proud and symbolic day for all Africans, whether resident on the continent or part of the vast diaspora. It is a celebration of unity, resilience and rich cultural diversity. It is also a reminder of the persistent challenges – poverty, inequality, unemployment, access to healthcare and education – that still affect so many.
Organisations and governments across the continent are utilising rapid technological advances and powerful AI-enabled innovations to improve the lives of all who call Africa home. Africa’s digital transformation is in full swing.
But there are also emerging risks, especially with AI, the seminal technology of our time. The global AI narrative often excludes African languages, customs and cultural contexts. This is not only a significant missed opportunity, but also a real risk: without intentional inclusion, Africa’s rich diversity is at risk of being overlooked and could be lost forever in the fast-evolving digital age.
Africa’s Rich Tapestry
Africa consists of 54 countries and is home to 1.4 billion people – nearly 20% of the world’s population. More than 3000 distinct cultures speaking over 2000 languages call Africa home. To put this into perspective, Africa has double the population of Europe and more than ten times the number of languages, highlighting the continent’s immense cultural and linguistic diversity.
And yet this vast diversity is at risk of being lost in the Age of AI. Most generative AI models are trained on Western data sets, predominantly in English and other dominant global languages. As a result, these technologies often fail to accurately interpret African contexts, leading to the misrepresentation of names, accents, traditions, and everyday experiences of a fifth of the global population.
According to the US-based public policy think tank Brookings Institution, Generative AI tools are often trained on internet data, meaning access to these tools may be limited to those who speak a few data-rich languages like English, Spanish, and Mandarin.
Localisation Matters
In the context of Africa and AI, localisation is essential for both inclusion and relevance. When AI systems fail to understand the nuances of local languages and cultural norms, they risk excluding entire communities from the digital revolution.
A clear example of the need for greater localisation can be found in the financial sector. The absence of financial education tools and banking apps in indigenous languages remains a major barrier to financial inclusion for hundreds of millions across the continent.
A research paper on Language Barriers in Financial Inclusion emphasized that these language challenges significantly hinder access to financial services, noting that many digital platforms are not available in local languages. The study identified the lack of localisation in financial tools as a primary factor contributing to the systemic exclusion of non-English-speaking communities from essential financial services.
Equally concerning is the potential erosion or complete loss of languages and traditions, many of which are already under serious threat.
According to the United Nations Educational, Scientific and Cultural Organization (UNESCO), Africa is home to one-third of the world’s languages. Alarmingly, 15% of these are at risk of extinction. These include the Khoi and San languages in Southern Africa, El Molo in Kenya, which is spoken by fewer than 500 people, and Bangi Me in Cameroon, which is spoken by fewer than 50 people.
In such extreme cases, generative AI could play a vital role in documenting, preserving, and even teaching these endangered languages to younger generations, helping to safeguard a critical part of our continent’s cultural heritage.
Supporting African AI innovation
We must acknowledge the encouraging rise of African-led innovation in this space, particularly the development of AI models trained in widely spoken African languages such as Swahili and isiZulu. We are also seeing the emergence of voice recognition tools tailored to African dialects and accents, as well as creative uses of AI to tell African stories and reflect local experiences. While these developments are promising, much work still lies ahead.
African governments, academic institutions, and global technology companies all have a critical role to play in accelerating this progress. In multilingual countries – which are commonplace across Africa – AI and technology can be powerful tools for closing communication gaps and enhancing inclusion and access.
The United Nations Development Programme (UNDP) highlights the importance of co-creating AI solutions that are representative of local priorities. It advocates for Africa to be not just a beneficiary, but an active partner in shaping the future of AI, ensuring that technological advancements align with the continent’s unique needs and development goals.
Localising tech is not just about translation, it’s about transformation. It’s about building digital tools and AI models that reflect African identities, cultures and values. It’s about ensuring that Africa isn’t just a consumer of global innovation, but a co-creator and co-innovator of it. In this regard, everyone needs to play a role, and the importance of Public-Private Partnerships cannot be emphasised enough.
On this Africa Day, let’s commit to shaping a digital future that doesn’t just include Africa, but is deeply and intentionally influenced by it. Let’s build AI that doesn’t just speak to us, but speaks like us — in our languages, through our stories, and with our spirit.
Dumisani Moyo is the Marketing Director at SAP Africa
