DIGITAfrica at Deep Learning Indaba 2026!
From 2–7 August 2026, Deep Learning Indaba 2026 took place in Lagos, Nigeria, bringing together members of the African AI community. DIGITAfrica partners from UCT, INRIA, UMA and UN participated in the event by hosting a dedicated workshop focused on the DIGITAfrica Blueprints.
The workshop provided an opportunity to present DIGITAfrica’s work on digital research infrastructure, with a particular focus on Connectivity and Edge AI and on approaches designed to support decentralised, privacy-preserving and sustainable infrastructure adapted to the African context.
Exploring the DIGITAfrica Blueprints
A central part of the workshop was the presentation of the DIGITAfrica 5-Tier Blueprints, including the Connectivity and Edge AI blueprints. These provide actionable frameworks and technical architectures for developing digital science infrastructures tailored to the African context.
The workshop also showcased progress on on-device Federated Learning (FL) models. Using this approach, edge devices such as off-the-shelf smartphones can collaboratively train machine-learning models locally without exposing users’ sensitive raw data.
Participants were introduced to several use cases covering AI, 5G and Smart Campus applications, demonstrating how the approaches developed through DIGITAfrica can be applied to different digital challenges.
Expanding Federated Learning Beyond Agriculture
The workshop also highlighted DIGITaAfrica’s work on Federated Learning in precision agriculture and crop disease detection.
Building on these initial implementations, DIGITaAfrica is looking to adapt and extend Federated Learning models to additional sectors. Potential areas include healthcare and diagnostics, climate and environmental monitoring, and local knowledge preservation and speech systems.
These future applications open opportunities for further collaboration between researchers, developers and domain experts interested in applying privacy-preserving AI approaches to new challenges.
The participation of DIGITAfrica partners at Deep Learning Indaba 2026 provided an opportunity to share the project’s work and engage with participants interested in the development of digital technologies and infrastructure for Africa.
The DIGITAfrica team thanks the Deep Learning Indaba 2026 workshop organisers, participants and consortium partners for their contributions and engagement throughout the workshop.
Join the DIGITAfrica Community!
Building sustainable, sovereign, and privacy-first AI infrastructure requires collective action. Whether you are a researcher, developer, domain expert, or student, we invite you to collaborate with us.
Get involved and join the DIGITAfrica Community of Practice here.
