At the Digital Africa Summit South Africa 2025, a high-stakes conversation unfolded on what it will take for Africa to build an AI future rooted in sovereignty, inclusivity, and local relevance. Moderating the session titled “AI for Good Governance: Data-Driven Decisions for Africa’s Future,” I guided a panel featuring Elizabeth Migwalla, Vice President of International Government Affairs at Qualcomm; Stefan Steffen, Technical Director of Data and AI at Cenfri; and Anthony Mveyange, Director of Program Synergy at APHRC.
While global AI forums often default to buzzwords and abstract optimism, this session focused squarely on practical pathways for Africa to shape its own technological trajectory. The discussion made clear that Africa’s ability to leapfrog depends not only on tools and algorithms, but on political will, shared data ecosystems, and a shift in mindset about the role of African languages and identities in the digital future.
Local Data, Local Languages, Local Futures
The conversation opened with a fundamental challenge: identifying where AI can meaningfully strengthen public-sector decision-making. Anthony Mveyange emphasized that Africa’s greatest opportunities lie in integrating locally relevant data to improve health, agriculture, and education systems. He argued that African languages must be central to innovation, noting that “innovation is born out of languages.”
With millions speaking Swahili, Zulu, Hausa, or Fulani, AI systems must be built to communicate in vernaculars that reach people where they are—including children and rural communities.
Africa’s linguistic diversity, he stressed, is not a burden but an advantage. “Our diversity is not a curse; it is an asset,” Mveyange said, urging leaders to abandon the idea that English is the sole language of innovation.
The Scalability Gap
For Stefan Steffen, one of the continent’s biggest barriers is fragmentation. He noted that countries are piloting similar AI applications—in agriculture, financial inclusion, and social protection—without shared frameworks or reusable platforms.
“We see AI advisory tools being built in Rwanda, Tanzania, Ethiopia, Ghana… but not as a shared conversation,” Steffen said. The solution, he argued, lies in modular, reusable assets such as small language models that operate offline on low-cost devices, enabling equitable access across markets.
Scaling these tools requires deliberate partnerships across public institutions, private innovators, and regional bodies.
Partnerships and Policy as Catalysts
Elizabeth Migwalla highlighted Qualcomm’s work putting edge AI capabilities directly into the hands of young African innovators. She positioned AI adoption not as a purely technical shift but an ecosystem challenge requiring technologists, regulators, entrepreneurs, and educators.
“It takes the intelligent youth—60 percent of our population. It takes technologists. It takes policymakers. It’s that kind of partnership that will close usage gaps,” she said.
But she warned that Africa risks deepening existing divides unless infrastructure, connectivity, and policy frameworks evolve alongside innovation. AI, she argued, must be understood not as a monolith, but as an enabler—helping citizens perform everyday tasks more efficiently, from farming to trade.
Trust, Transparency, and Citizen Literacy
As AI tools begin to influence public service delivery, trust and accountability become non-negotiable. Steffen proposed a concrete transparency model: an algorithmic “QR code” that citizens can scan to see what data was used, who built the model, and how to challenge inaccurate decisions.
He also emphasized that digital and data literacy programs must extend to both citizens and national parliaments. Cenfri’s work with Rwanda’s legislature, he noted, equips leaders to “lead from the front” on AI policy decisions.
When AI is well implemented, he added, it becomes invisible—citing Google Maps as an example of sophisticated AI that users scarcely recognize as such.
Africa’s Seat at the Global Table
The panel converged on the geopolitical stakes of AI development. Migwalla warned that Africa must help define global rules, not simply absorb them. With developed economies investing more in AI than many African nations’ GDPs, the continent risks being shaped by external priorities unless it asserts its sovereignty.
Her message was blunt: “If we are not at the table creating the directions, then we are on the menu.”
Prioritizing Impact Sectors
When asked to identify priority sectors for AI deployment, the panelists pointed to foundational areas:
- Health and Education – Mveyange emphasized education for “epistemological transformation,” enabling societies to rethink how knowledge is produced and applied.
- MSMEs – Steffen highlighted small enterprises as engines of jobs and growth, where AI can reduce costs and improve market access.
- Agriculture – Migwalla stressed that food security is existential: “If you can’t eat, if you can’t produce food,” innovation elsewhere becomes irrelevant.
Conclusion: Toward a Unified African AI Agenda
The session affirmed that Africa’s AI debate has moved past speculation into urgent strategy. The priorities are clear: build shared data ecosystems, design scalable and reusable models, invest in transparency, and develop language-centric AI systems that reflect Africa’s cultural and linguistic richness.
The path forward demands collaboration across borders, sectors, and disciplines—and a firm belief that Africa’s diversity is a catalyst for innovation, not a constraint.
If African leaders, innovators, and citizens can embrace this collective agenda, AI can evolve into a force for equity, accountable governance, and inclusive growth. As the discussion revealed, the future will not be shaped by code alone, but by continental cooperation, intentional policy, and an unwavering commitment to African-centred design.
