Mapping the AI narrative in Kenya and South Africa's media

Mapping the AI narrative in Kenya and South Africa's media A new study by the Centre for Information Integrity in Africa (CINIA) at Stellenbosch University, supported by DW Akademie, takes a critical look at how artificial intelligence (AI) is covered in Kenyan and South African media. The research, titled "AI in the Media: Kenya and South Africa", analyses 57 articles and draws on two rounds of journalist interviews over a five-year period (2021-2026). The research finds that AI remains largely an underexplored, and often misunderstood, topic for journalism in South Africa and Kenya. Here are the key takeaways. Siloed and reactive reporting AI articles tend to be siloed and reactive, framed as business, security or technology stories. Seldom is AI presented in terms of its multidimensional, layered impact on society. Furthermore, the geopolitical aspect of AI rarely gets interrogated in mainstream media. Coverage is often triggered by crises or scandals rather than proactive public-interest journalism. Eight key themes Eight substantive themes emerge from the research: AI for development, inclusion and social change; AI as a tool for disinformation and political manipulation; data colonialism and infrastructure gaps; AI governance, ethics and policy; AI and workforce/labour exploitation; AI's impact on journalism and the media; authoritarian tech and crime; and environmental concerns. 1. AI for development, inclusion and social change Optimistic narratives in African media coverage of AI frequently emphasise the technology's potential to support development, education, healthcare, and social inclusion. Rather than framing AI primarily as a disruptive or threatening technology, these accounts present it as a tool capable of addressing structural challenges such as educational inequality, skills shortages, limited access to services, and financial exclusion. 2. AI as a tool for disinformation and political manipulation Alongside optimistic narratives about innovation and development, African media coverage of artificial intelligence increasingly highlights concerns about the technology’s role in disinformation, political manipulation, fraud, and state surveillance. In this framing, AI is often portrayed as a double-edged tool: capable of improving governance and administrative efficiency, but also vulnerable to misuse by political actors, cybercriminals, and governments seeking greater control over digital spaces. 3. Data colonialism and infrastructure gaps Debates about data colonialism reflect the power dynamics behind tech and AI in particular, and touch on issues of data extraction from less developed nations by tech companies located predominantly in China and the Global North. They reflect issues of exploitation and privacy, and many draw on parallels with historic narratives of colonial extraction. 4. AI governance, ethics and policy This is a theme that seeks to shed light on the rules, ethics and practice of AI and how it shapes society. While such stories can often feel quite abstract, focusing on the ecosystem in which AI operates, some reporting puts flesh on the bones with practical real-world examples. 5. AI and workforce issues, labour exploitation Workforce issues and AI have tended to revolve around job displacement narratives, including in the field of journalism. As AI tools undertake tasks that humans did previously, and skills, readiness and training, a disproportionate number of texts focusing on the media sector and job displacement. Numerous interviewees referred to editors as "technophobic", with older journalists seemingly resistant to embracing new technologies. 6. AI's impact on journalism and the media African media debates increasingly highlight how artificial intelligence is reshaping journalism, both by offering new tools for reporting and by creating new risks for press freedom, information integrity, and newsroom sustainability. Generative AI can produce deepfakes and large volumes of synthetic content, while algorithmic systems on digital platforms amplify misleading narratives. These developments complicate journalists' ability to verify information and maintain credibility in rapidly evolving digital information ecosystems. At the same time, journalists are exploring how AI can support newsroom practices, for instance analysing large datasets, summarising research, and organising investigative materials. 7. Authoritarian tech and crime Authoritarian tech is broadly described as the use of modern technologies, especially digital tools like AI, surveillance systems, and internet controls, by governments to monitor, manipulate, and suppress populations. Its founding principle is that technology is not neutral, and can be weaponised by those who control it. The intersection of crime and mass surveillance has found traction in both Kenya and South Africa. 8. AI and the environment A newly emerging topic of interest to journalists is the environmental impact of massive AI expansion. It feeds into the debate about infrastructure, regulation and the opportunity cost of Africa being a key player in the AI global economy and the impact that may have on limited resources. African companies are joining the global rush to develop AI infrastructure. This will require investment in data centres, which are known to have a significant impact on the environment as they require land to be made available to house them, electricity to run them, and water to cool them down. Vendor narratives and the knowledge gap Given the apparent knowledge gap about AI among senior editorial staff, tech companies are in a position to fill that void, supplying copy to journalists that positions AI in a favourable light. Structural and professional barriers were also noted. Journalists cite technophobia among senior editors, fear of AI-driven job displacement, newsroom under-resourcing, and story competition from pressing social issues as reasons why AI coverage remains shallow and hard to commission. Global North framing dominates The topic of AI is often presented through a Global North lens, with little contextualisation or interrogation of these issues from an African point of view amid resource and access constraints. This occurs against a background of under-developed policy frameworks for AI on the continent generally, and in newsrooms in particular. Coverage frequently mirrors international narratives about the AI arms race between the US and China, with insufficient contextualisation for African realities, including digital exclusion, infrastructure deficits, and language marginalisation. Recommendations The report makes several recommendations to improve coverage of AI in Africa. Practical assistance may include: - Fellowships and funding opportunities for journalists dedicated to the coverage of AI, and for investigative journalism projects on AI. This would also contribute to more in-depth journalism on the topic. - Cross-border journalism initiatives should be supported to enable closer monitoring of AI reporting in Africa. It also provides an opportunity for joint investigations to be undertaken and potentially offers a global platform to showcase the best in African AI journalism. - Development of an AI reporting toolkit for journalists in Africa. This could act as an AI literacy resource as well as a tool to guide journalists towards the key thematic areas to consider when preparing a story on AI. The toolkit could include a practical newsroom AI reporting manual with glossaries, key thematic areas and angles to consider. - In-newsroom training on reporting AI could be integrated into broader training being undertaken on the operational use of AI in newsrooms. The training must be locally relevant and practically focused. - Showcasing the best AI stories from the region, similar to the World Press Photo Contest for photojournalism, could help give the subject public prominence. Karen Allen is a former BBC correspondent with 25 years of experience covering Africa and the Middle East. She is now an independent consultant advising organisations globally on digital transformation in journalism, information integrity and emerging threats. She has undertaken studies on the impact of information interference and manipulation on elections in Kenya, South Africa and Uganda and consults for the Institute for Security Studies, Centre for Information Integrity in Africa (CINIA) and the Konrad Adenauer Stiftung. Allen holds a Master’s degree in International Relations and Contemporary War from King’s College London and is a visiting scholar in the Department of War Studies. She is also a Fellow at Stellenbosch University and a senior researcher at CINIA. Herman Wasserman is Professor of Journalism and Director of the Centre for Information Integrity in Africa at Stellenbosch University. He worked as a newspaper journalist before embarking on an academic career. He previously taught at the universities of Sheffield, Newcastle, Rhodes and Cape Town, and has held several visiting professorships, including at the Ludwig Maximilian University (Munich). He has published widely on journalism ethics, media, conflict and democracy and disinformation in Africa. He has consulted for, among others, OpenAI, UNESCO, DW Akademie and Reporters Without Borders.