Keynotes & panels

As artificial intelligence becomes central to engineering practice, its teaching remains largely shaped by assumptions of abundant resources, high-performance computing, large datasets, and persistent connectivity. However, in many parts of Africa and other resource-constrained regions, engineers must design systems under radically different conditions.

This keynote argues for a paradigm shift in engineering education: from resource-intensive AI training to constraint-driven, frugal, and context-aware learning models. It introduces the concept of “frugal AI education”, where students learn to design intelligent systems using low-cost hardware, limited data, and energy-efficient computation.

The talk explores how emerging approaches such as embedded AI, edge computing, and distributed systems can be integrated into engineering curricula through project-based learning, open hardware platforms, and community-driven innovation labs. It also highlights the importance of aligning pedagogy with local challenges, particularly in agriculture, environment, and infrastructure. Beyond technical skills, the keynote emphasises the need to cultivate engineers capable of co-designing with communities, maintaining systems locally, and innovating under constraint. It ultimately proposes a vision for engineering education that is inclusive, resilient, and rooted in local ecosystems, preparing a new generation of engineers to build sustainable AI solutions where they are needed most.

Continental prosperity is the vision of an economically integrated, developed, and stable continent. It encompasses a high standard of living, poverty reduction, job creation, improved education and healthcare, and sustainable development for all citizens across the continent. Engineering education is a catalyst for change. It promotes national and regional prosperity by developing skills to create innovative products and processes, drive economic growth, solve societal challenges in sustainable ways, and foster technological advancement.

The continent of Africa can achieve immense prosperity through engineering education if member nations, individually and collaboratively, make strategic investments in engineering capacity and infrastructure, which constitute the most direct path to national or continental prosperity. Artificial intelligence (AI) plays a crucial role in driving continental prosperity through engineering education in different ways. These include enhancing education systems, shaping the future workforce, and stimulating innovation. AI can also contribute significantly to economic growth, technological advancement, and overall societal well-being.

Resource-constrained environments like ours are characterized by limited laboratory infrastructure, unreliable internet access, a low ratio of teachers to learners, low parental literacy and restricted access to updated learning materials, and face systemic barriers to producing and transferring knowledge. This talk examines the transformative potential of Artificial Intelligence in reshaping pedagogical practices and operational models across resource-constrained educational settings.

Firstly, it focuses on the key challenges that AI models face, including data poverty, algorithmic bias in low-resource contexts, lack of infrastructure, lack of teacher preparedness, financial constraints and language barriers. Secondly, while taking into account the realities of the local context, the talk explores how some AI frameworks can help to mitigate these challenges: frugal AI for resource reduction, adaptive tutoring systems for personalizing learning pathways, generative AI for simulation, and AI-assisted assessment and feedback for reducing faculty workload. Finally, it presents a few ideas demonstrating how AI can facilitate the democratization of education, shifting the focus from passive content absorption to active, iterative and ethical design-based learning — from exclusive to inclusive education.

Artificial intelligence is reshaping journalistic practices, from automated content and fact-checking to data analysis. How, then, should journalism education evolve to prepare students for newsrooms that are increasingly augmented by AI? This contribution explores three pedagogical axes. First, AI as a learning tool: students can use text generators and verification assistants to gain efficiency while developing a critical eye toward their biases and limitations. Second, AI as an object of investigation: training future journalists to decode the algorithms behind news recommendations, deepfakes, and content personalization, thereby strengthening their digital literacy. Third, AI as a pedagogical simulator: chatbots can act as sources, opponents, or audiences, enabling realistic role-playing scenarios and individualized feedback.

Ethical questions are also addressed: how far can AI assist without replacing human judgment? How can dependency be avoided while preserving the ethics of verification? Feedback from a master’s program in journalism illustrates both benefits (time saved on repetitive tasks, access to large datasets) and necessary precautions (transparency, supervision, source diversity). The ambition is not to mold technicians of AI, but to cultivate journalists capable of engaging with these tools while keeping their professional sovereignty intact. To collaborate, yes — but without relinquishing agency.

Writing reports, preparing teaching materials, checking the treatment protocol for a disease, farming specific crops, even looking for a husband or wife: all these activities are now entrusted to AI, and we hope and trust AI to deliver a quick and best reply. There is a need for frugal, responsible and resilient use of AI tools. Nowadays, people tend to be more willing to trust AI tools than to interact with humans.

Defining a framework and training teachers and learners are therefore necessary. This contribution addresses the baseline for a responsible use of AI tools in education, and the urgent need to empower and train teachers and learners by making them more responsible and creative in the age of AI.

Artificial Intelligence is reshaping engineering education globally. Yet its integration in African contexts must be approached responsibly, contextually, and sovereignly. AI systems learn from documented and digitized histories; when local knowledge remains orally transmitted or underrepresented in global datasets, risks of epistemic invisibility and bias amplification emerge. At the same time, infrastructural constraints, economic dependency, and governance gaps shape the feasibility and direction of AI adoption.

This keynote demystifies the technical foundations of AI, clarifying how data distributions, optimization objectives, and deployment conditions influence model behavior. It examines structural constraints specific to resource-limited environments and analyzes systemic risks operating at algorithmic, cognitive, institutional, and epistemic levels. Particular attention is given to the role of Digital Humanities in documenting and structuring locally grounded engineering knowledge, thereby expanding the representational foundations upon which AI systems learn.

Rather than advocating uncritical adoption, the keynote argues for a balanced integration strategy that builds upon strong scientific foundations already present in engineering curricula. Responsible and sovereign AI ecosystems cannot rely solely on technological advances: they require sustained collaboration between engineering sciences, social sciences, humanities, and policy actors on governance, inclusion, cultural representation, and societal impact — supported by funding initiatives that enable interdisciplinary research, capacity building, infrastructure and regional cooperation aligned with African realities and aspirations.