Abstract
Background: Generative artificial intelligence is now part of the ordinary ecology of higher education, but institutional governance has not matured at the same speed as tool adoption. Purpose: This article examines how generative AI integration, policy visibility, readiness, opportunities, and risks are publicly documented at the African Leadership University (ALU) and the Adventist University of Central Africa (AUCA) in Rwanda. Method: The study uses an interpretivist comparative multiple-case design based on public document analysis. The core corpus comprises publicly retrievable institutional webpages, policy documents, academic regulations, handbooks, e-learning materials, research manuals, national policy texts, and sectoral guidance retrieved and checked through April 2026. Analysis used a directed qualitative codebook aligned with institutional readiness, Diffusion of Innovations, and a rights-based AI governance lens. Findings: Both universities publicly signal AI engagement, but in different forms. ALU is more innovation-signaling, with visible AI research, student bootcamp activity, and outward-facing policy discourse. AUCA is more governance-dense, with stronger public visibility of academic regulations, ICT and online-learning policies, plagiarism infrastructure, and AI-and-big-data institutional positioning. Yet neither university publicly presents, within the reviewed corpus, a fully specified institution-level generative AI acceptable-use architecture aligned with Rwanda’s emerging sectoral expectations. Contribution: The article sharpens a distinction between visible innovation activity and visible governance readiness, showing why AI adoption cannot be inferred from promotional visibility alone. It also offers a replicable public-document method for studying institutional AI governance in underdocumented African higher education contexts while clearly separating documented evidence, analytic inference, and public silence.
Showing the abstract — retrieve the full paper via the Exa API.