For nearly a decade, Prof. Vukosi Marivate has been working to ensure that Africa shapes artificial intelligence rather than being shaped by it. His appointment to the United Nations Independent International Scientific Panel on Artificial Intelligence — the first global scientific body of its kind — has given that mission an international platform.
Selected from more than 2,600 applicants across 140 countries, Marivate is one of 40 experts chosen to sit on the panel. He serves as director of the African Institute for Data Science and Artificial Intelligence at the University of Pretoria and holds the ABSA UP Chair of Data Science.
“My personal mission is to make sure that Africans have agency during these revolutions,” Marivate told Forbes Africa. “If you are not actively working to build up your own capability and capacity, it is somebody else doing it on your behalf, for their own reasons.”
The appointment builds on years of institution-building. In 2017, Marivate co-founded the Deep Learning Indaba, widely regarded as the continent’s largest AI research conference. He also co-founded Masakhane, a grassroots research initiative with thousands of volunteers building natural language processing tools for African languages, and Lelapa AI, which has launched InkubaLM, described as Africa’s first multilingual large language model.
The question of who benefits from AI — and who is simply used by it — runs through all of his work. “We don’t want to be at the table, being on the menu,” he said. “It is very important to have these technologies really represent people, because from there they can actually reach the outcomes we think AI is going to deliver. Just by saying AI exists, it is not going to solve education.”
South Africa has 12 official languages, yet most AI-powered voice assistants speak none of them. Marivate said the challenge of representing African languages in AI goes well beyond missing datasets. “Historical, cultural and political factors have influenced how languages were represented in digital systems,” he said.
The UN panel will deliver its first report at the Global Dialogue on AI Governance in Geneva in July 2026, with a mandate to ground international AI policy in scientific evidence rather than speculation. Marivate said he does not expect a single global framework to emerge from the process, but believes broadening the range of voices shaping AI governance is itself a significant outcome. “Decisions about artificial intelligence have often been driven from only one or a few worldviews,” he said. “The reason for this panel is to bring in more voices and perspectives about what AI will mean for the world.”
He warned that Africa’s under-investment in research and development has left it poorly positioned to influence the AI revolution, and that the lesson must be learned before the next transformative technology arrives. “If everything just accrues in one space, in one place, we are just spectators,” he said. “If we’re not doing that, we are not in control of whatever our future is going to be.”
