Only a few years ago, AI policy on the continent revolved around ethics, digital literacy and startup incubation. Now governments are discussing cloud infrastructure, sovereign data, regional computing capacity and local language models, subjects once confined to engineers and Silicon Valley executives.