The Deep Learning Indaba is, and always will be, its people [post Deep Learning Indaba 2026]

During the Mutapa Empire, which encompassed parts of what are now Zimbabwe and Mozambique, the royal court, the dzimbahwe, moved from place to place rather than remaining in a single location. As it travelled, it brought information to communities, including news, updates on governance, agricultural knowledge and advice, as well as opportunities for the exchange of ideas. In this way, the movement of the royal court helped the empire to grow, learn and innovate. It also created opportunities for local traditions to interact with and contribute to a shared identity across the empire. A similar tradition existed in West and Central Africa. During the Mali Empire and the subsequent Songhai Empire, griots, living historians somewhat akin to imbongi in Southern Africa, travelled between kingdoms, gathering communities, preserving local histories and practices, and sharing knowledge. This heritage of movement, knowledge-sharing and exchange across Africa, documented beautifully in Zeinab Badawi‘s 20-part BBC documentary series, The History of Africa, and her book, An African History of Africa: From the Dawn of Humanity to Independence, provides an apt way for me to explain what the Deep Learning Indaba is and why it moves from country to country every year.

The Deep Learning Indaba (DLI) is an annual gathering that brings together researchers, developers, students, practitioners and communities working in Artificial Intelligence (AI), Machine Learning (ML) and Data Science (DS) across Africa. The word “indaba” refers to an important community meeting in isiZulu, and the name captures something fundamental about the event. It is a space where people come together to exchange ideas, challenge one another and collectively consider the future of AI. Since its first gathering in 2017, the DLI has grown into the largest gathering of the African AI, ML and DS research, development and innovation community. The annual gathering is part of a much larger network of local IndabaX communities, our “kingdoms”, which now span 47 of Africa’s 55 countries. This is a remarkable achievement that should not be overlooked, and much credit is due to the IndabaX teams that have built and sustained these communities. These connect with other grassroots orgs like Data Science Nigeria, Data Science Africa and more.

The volunteers who make the DLI possible have created a space where people from across the continent can share knowledge, challenge one another and develop new ideas. Its purpose is not simply to showcase advances in AI and ML, but to ensure that Africans are actively involved in shaping the technologies and research agendas that will affect the continent. For me, this has always been what makes the DLI special. It is ultimately a community built around its people and the relationships they create.

Attendees are often struck by the intensity of the week and the sheer amount of exchange that takes place. The intensity comes from everything that happens within that single week. Organisers and participants move between research presentations, workshops, panels, mentoring sessions and informal conversations, sharing knowledge openly and challenging one another’s ideas. Difficult and sometimes controversial questions are debated, ideas are tested, and perspectives from different disciplines and parts of the continent come together. The pace can be demanding, but it is also what makes the Indaba so productive. At its heart is a shared commitment to ensuring that Africans are not simply consumers of AI, but are actively shaping its research, development, infrastructure and use on the continent.

I have had the privilege of being part of this journey from the beginning as one of the co-founders of the DLI. The first gathering took place in 2017 at the University of the Witwatersrand in Johannesburg, South Africa, where I served as a co-founder/organiser. I then served as a member of the DLI board until December 2025. What began as a gathering of people with a shared interest in building an African AI community has since grown into a continent-wide network of researchers, developers, students, practitioners and local communities. It has been remarkable to see how far the community has come and how many people have contributed to its growth.

The 2026 gathering was another important milestone for the community. Held during the first week of August at Pan-Atlantic University in Lagos, Nigeria, the gathering brought together an exceptionally diverse group of people working across AI, ML and Data Science. I attended as an invited keynote speaker, which aligned closely with the focus of my 2026 sabbatical on improving evaluations for African-language NLP systems. I also had the opportunity to contribute through panel discussions and both formal and informal mentoring. Having been involved with the DLI since its beginnings, it was particularly rewarding to see the community continue to develop while retaining the openness and sense of shared purpose that have always characterised the Indaba.

I particularly enjoyed the plenary keynotes by Catherine Nekalembe, who introduced a term that neatly encapsulated some of the work we have been doing at DSFSI, Pasteur’s Quadrant, and Margaret Mitchell, who consistently challenges us to examine what we are building, why we are building it and how we arrived at where we are today. The poster sessions, publication track, workshops and panel discussions provided further opportunities for exchange. Throughout the programme, the strength of the African AI community was evident in the way researchers and practitioners connected their work to local needs, challenges and opportunities. The programme brought together cutting-edge research and innovation across disciplines, and the depth and breadth of the work presented were remarkable.

This year’s programme reflected the increasingly important questions facing African AI. Discussions ranged from compute sovereignty and responsible and sustainable data gathering to efficient methodologies, AI governance, AI safety and GeoAI, among many other topics. These conversations demonstrate that the DLI is not simply following global discussions about AI. It is creating space for African researchers and communities to define the questions that matter to the continent and to contribute to the development of solutions.

The DLI’s strength also lies in its ability to move across the continent while remaining connected to the communities it serves. We experienced Nigeria this year, Rwanda in 2025, Senegal in 2024, and other countries in previous years. Each host brings its own context, priorities and communities to the Indaba, while the wider DLI community carries knowledge and relationships from one gathering to the next. In this sense, the Indaba is both a travelling conference and a growing continental community.

The DLI’s journey will continue. The 2027 edition will be an important occasion as the Indaba returns to South Africa for its 10th anniversary. Looking further ahead, the planned transition to a new model in 2028, with countries bidding to host future Indabas, will open another chapter in the community’s development. Having watched the DLI grow from its first gathering in Johannesburg into a continent-wide movement, I am particularly excited to see where it goes next.

Prof. Vukosi Marivate is Director of the African Institute of Data Science and Artificial Intelligence (AfriDSAI), ABSA UP  Chair of Data Science and Professor of Computer Science at the University of Pretoria, where he leads the Data Science for Social Impact group. His research focuses on Machine Learning (ML), Artificial Intelligence (AI), and Natural Language Processing (NLP), particularly for African and other low-resource languages. He co-founded Lelapa AI, the Masakhane Research Foundation, and the Deep Learning Indaba. His work spans social challenges in science, energy, public safety, and utilities, aiming to create AI for Africans by Africans. Vukosi is a member of the Independent Scientific Panel for AI for the UN and a member of the African AI Council at Smart Africa.

Thank you to Dr Angel Maduke for editing and refining the piece.

Prof Marivate’s Sabbatical is supported by Oppenheimer Memorial Trust.

A beautiful exhibit at the Yemisi Shyllon Museum of Art at the Pan Atlantic University campus.