Date: Monday, August 24, 2026
Hello! I am Varaidzo Magodo-Matimba, a core collaborator at The MERL Tech Initiative and the AI+ Africa Learning Group Lead, hosted by the Natural Language Processing Community of Practice. The AI+ Africa Learning group exists to create space, structure and momentum to work and learn together. We have over 700 members and our membership spans multiple sectors and we work to build relationships and connections across and within the continent, with an eye toward strengthening Majority World collaborations.
The reflections I will share in this post are musings of practical exploration and embodiment of Made in Africa AI approaches in evaluation and what our Made in Africa Landscape Study from 2025 surfaced.
1. What AI in Evaluation Means for African Evaluators – The rise of AI in evaluation calls for three core competencies. Together they form a simple sequence: ask hard questions beyond the technical, push for tools that fit the African context, and keep sight of the bigger picture.
The first is critical and political awareness. African evaluators need to ask not just whether an AI tool works, but whose assumptions and interests it carries. That matters, because development aid has a long track record of reinforcing colonial narratives especially in evaluation methodologies.
Second comes contextual and technical fluency: the skill to assess, adapt, and help design AI tools, including small African language models, that reflect indigenous knowledge, and lived realities.
Third comes systems and power analysis. Closing the AI gap is about reshaping who controls and owns the technology, data, and knowledge. Together, these competencies help evaluators hold AI accountable, partner with communities, and make sure technology strengthens evaluation across Africa.
2. Interrogation by African evaluators is the “word of the day” – So what does this look like in practice? Right now, the African evaluator’s role is not to adopt or reject AI only, but to interrogate it. We interrogate because evaluators have always shaped what counts as evidence, whose voice matters, and which outcomes get called a success. We question because most commercial AI runs on English-language data and Western frameworks. As a result, oral traditions, local languages, and community knowledge can slip through the cracks unless evaluators actively protect them. We interrogate because good intentions are not enough. Movements like Made in Africa Evaluation give us the values: African agency, ownership, and context-driven ethics. Yet the Made in Africa AI in MERL Landscape Study shows the harder truth that practically embodying those values takes steady, deliberate effort.
3. Evaluators Are Central to Realizing Made in Africa AI Approaches – This is why evaluators matter as the AI trajectory in Africa develops. Evaluators need to call out “AI inevitability” versus “shaping AI in African evaluation “. It is easy to treat AI as an unstoppable force. AI in Africa results from real choices made by tech companies, governments, donors, and local actors, each with “clashing agendas”.
Evaluators see this clearly and can push back when AI is imposed rather than co-created. More than that, they must actively demand African-led design and governance but setting the metrics of applicability and contextualization in AI deployed in Africa. Here, “Made in Africa AI in evaluation” becomes real, and decolonial evaluation leadership takes shape.
Ready to test your knowledge? I recommend that you audit your skills across the seven Made in Africa AI in MERL competencies with our self navigation framework!
Do you have questions, concerns, kudos, or content to extend this aea365 contribution? Please add them in the comments section for this post on the aea365 webpage so that we may enrich our community of practice. Would you like to submit an aea365 Tip? Please do so using the form on our website. aea365 is sponsored by the American Evaluation Association and provides a Tip-a-Day by and for evaluators. The views and opinions expressed on the AEA365 blog are solely those of the original authors and other contributors. These views and opinions do not necessarily represent those of the American Evaluation Association, and/or any/all contributors to this site.