Automating or Preventing Inequality: Will AI Solve or Cement Inequality and Poverty Across Africa?

Sub-Saharan Africa stands at a crossroads. By 2030, the region is projected to be home to 90% of the world's extreme poor and whether AI becomes a tool for transformation or a driver of deeper inequality may determine the trajectory of millions of lives. AI carries dual potential: it can make public services more equitable and accessible, while also amplifying misinformation and enabling identity-based violence. Without deliberate intervention, the risk is a future where AI automates poverty rather than solving it entrenching algorithmic inequality and creating new forms of exclusion that are invisible to policymakers. The benefits of AI must not accrue only to those with the infrastructure and digital fluency to access them. The choices made today will shape whether AI lifts the region's most vulnerable communities or leaves them further behind. To help close these critical knowledge gaps, the International Development Research Centre (IDRC) and the UK’s Foreign, Commonwealth and Development Office (FCDO), through the Artificial Intelligence for Development (AI4D) programme, have launched a research initiative to generate actionable, locally grounded evidence. Supported by diagnostic research from Genesis Analytics, the initiative advances a focused research agenda across four thematic areas. Poverty and inequality the focus of this brief is one of these priority areas, aimed at strengthening evidence and empowering African researchers to inform policies that shape the continent’s role in the global AI economy.