The Future of University

When people in Silicon Valley talk about the future of university education in the age of AI, they usually imagine campuses with hologram lecturers, AI tutors available 24/7 and students “co-creating” with algorithms.

From Johannesburg to Lagos to Kinshasa, the picture is more complicated. In many African universities, students still queue for hours to register on paper, share a single textbook between five, and study in dorm rooms where electricity cuts out at exam time. Yet the same students are already using ChatGPT on cheap smartphones over unstable connections.

The question for the Global South is not whether AI will transform post-secondary education. It is whether that transformation will narrow or widen the gaps — between rich and poor institutions, between the North and the South, and within African societies themselves.

1. A new wave of disruption on an old, uneven sea

For decades, policymakers have promised that digital technology would “leapfrog” structural problems in African higher education: overcrowded lecture halls, staff shortages, outdated curricula, brain drain. AI is the latest and most powerful version of that promise.

Generative AI tools are already reshaping universities worldwide — helping draft essays, summarize research, generate code and even provide instant feedback on student writing. A recent global survey across 76 countries found high awareness and widespread use of generative AI among students and faculty, though with strong regional and disciplinary differences. SpringerLink+1

A 2025 systematic review focused specifically on African Higher Education Institutions (AHEIs) reports the same pattern: huge interest in generative AI for teaching, learning and administration, but also distinctive constraints around infrastructure, staff training and ethical governance. SSRN+1

This is the paradox: Africa may have the most to gain from smart, low-cost educational technologies — and also the most to lose if AI arrives as yet another wave of digital colonialism, controlled from elsewhere, trained on other people’s languages and histories.

2. What AI could make possible for African universities

Measured against the realities of underfunded lecture halls and overloaded staff, the potential benefits of AI in African post-secondary education are striking.

2.1 Mass personalization in crowded systems

Across the continent, many universities are struggling with massification: huge growth in student numbers without matching investment in faculty or infrastructure. In such settings, the idea that each student could receive personalized feedback has long felt unrealistic.

AI-powered tools can, at least in theory, help:

• Automated tutoring systems can offer on-demand explanations, practice questions and tailored hints in subjects like math, programming, economics and statistics.
• Writing assistants can help students draft, structure and revise essays — particularly valuable where school systems have not prepared them for academic writing in English or French.
• Adaptive learning platforms can identify gaps in understanding and direct students to targeted exercises, allowing human lecturers to focus on higher-order questions rather than basic remediation.

An IDRC (International Development Research Centre) initiative on AI in African education notes that AI could support individualized learning, help students with disabilities, assist in first-language learning, and reduce administrative burden at school and university level. idrc-crdi.ca

If deployed with care, such tools could make crowded classes more humane — not by replacing lecturers, but by freeing them from some routine tasks so they can teach, mentor and research more effectively.

2.2 Teaching in African languages and across language barriers

Language remains one of the most underestimated barriers in African higher education. Many students are taught in English, French or Portuguese, even though they think, dream and argue in other languages.

AI-driven translation and speech technologies open new possibilities:

• A lecturer can speak in Kiswahili or Amharic while students receive real-time captions or translations into English for note-taking — and vice-versa.
• Course materials can be drafted in English but translated, with human editing, into isiZulu, Hausa or Kinyarwanda.
• Students can practice academic writing in their home language and then work with AI to translate and refine into the language of assessment.

The African Union’s Continental Artificial Intelligence Strategy, endorsed in 2024, explicitly recognizes AI’s potential to support multilingualism, preserve African cultural heritage and strengthen local knowledge systems, if appropriately governed. Union Africaine+2Union Africaine+2

In other words, AI could help African universities teach from Africa, not only about Africa in borrowed languages.

2.3 Research support and global visibility

For early-career scholars and graduate students, AI can act as a research assistant:

• Scanning large literatures and summarizing debates,
• Helping generate code for data analysis,
• Drafting outlines for grant proposals and articles.

This matters in contexts where access to expensive journal databases, research assistants or writing centres is limited. AI could help close the gap in research productivity and visibility between African universities and their northern counterparts — but only if African institutions also secure access to computing resources, data infrastructure and fair licensing.

UNESCO’s AI needs assessment survey in Africa points to encouraging signs: community-run AI classes on weekends, grassroots AI boot camps for young people, and university initiatives experimenting with machine learning for agriculture, health and education. UNESCO+1

The future of post-secondary education in Africa will be shaped by whether these grassroots energies connect with formal institutions and national strategies, or remain isolated pilots.

3. The structural obstacles: electricity, infrastructure, and inequality

For every hopeful scenario, there is a hard constraint.

3.1 The digital divide is not just about devices

AI systems are resource-hungry. They require reliable electricity, stable broadband, data centres and often cloud contracts denominated in dollars. Many African universities lack all four.

A 2024 Brookings analysis on AI in the Global South warns that inadequate infrastructure — especially unreliable electricity and limited internet access — could severely hinder AI development and deployment in these regions. Brookings

A Columbia University–linked report on AI-driven education adds that students in the Global South are systematically disadvantaged when AI tools assume uninterrupted connectivity and up-to-date hardware; even when tools are “free,” the cost of data and devices remains a barrier. Centre pour le Développement Durable

In this context, comparing AI policy in Ohio — where a major public university can decide to make AI training mandatory for all freshmen The Guardian — with policy in Kinshasa or Bamako is like comparing a high-speed train with a minibus stuck in traffic. They share the same name (“higher education”) but not the same starting point.

3.2 Whose data? Whose cloud?

Most of the powerful AI models used in classrooms today are built and hosted by a handful of companies in North America, Europe and East Asia. African universities rarely control the data, the source code or the servers.

This raises several concerns:

• Privacy and sovereignty: where do student essays, exam answers and research ideas go once they pass through commercially hosted AI systems?
• Cost and dependency: will African institutions become locked into subscription models they cannot shape and may not sustain?
• Content bias: how will African histories, literatures and social realities fare inside models trained predominantly on English-language, Global North data?

A NORRAG commentary on AI and unequal knowledge in the Global South argues that current AI systems risk reproducing old hierarchies: privileging Western epistemologies, marginalizing local knowledge and amplifying biases around race, gender and geography. Norrag Education

If African universities simply plug into imported AI systems without negotiating governance, there is a real risk that the continent’s post-secondary education becomes a consumer of other people’s knowledge architectures.

3.3 Gender and labour market risks

The implications of AI for post-secondary education go beyond lecture halls and libraries; they spill into the labour markets universities are supposed to prepare students for.

A 2025 study presented at the Global AI Summit for Africa in Kigali warns that women in Africa’s outsourcing sector are more likely than men to see their tasks automated by AI by 2030, with lower-paying roles particularly at risk. AP News

If universities and colleges don’t plan for this, AI-ready curricula could end up producing graduates for jobs that vanish, or reinforcing gender inequalities as “low-skill” tasks are automated and high-skill AI roles go to those who already had better schooling.

4. Case studies: Africa’s cautious experiments

Even within these constraints, African institutions and governments are not standing still. A picture of cautious, uneven experimentation is emerging.

4.1 South African universities: AI, decolonisation and curriculum reform

A recent article in Frontiers in Sociology examines AI in South African universities through the lens of curriculum transformation and decolonisation. It notes that while AI could help pluralize knowledge and support students from diverse backgrounds, it can also entrench Western-centric biases and exacerbate accessibility disparities if deployed uncritically. Frontiers

In practice, some South African campuses are exploring AI-assisted tools for academic writing support and coding, while student and staff debates focus on plagiarism, algorithmic bias and the risk that AI-generated content might dilute efforts to center African perspectives in the curriculum.

The core question is not simply “ban or allow AI?” but: Can AI be aligned with decolonial goals, or does it inevitably recentre external epistemic authorities?

4.2 Pan-African initiatives and teacher training

UNESCO’s regional work on AI and education systems in Africa has sought to frame AI not as a magic bullet but as one component in broader efforts toward inclusive, equitable, quality education. It emphasizes teacher training on AI competencies, ethical guidelines, and safeguards to protect human agency in an era of automation. Institut Afrique Capacités+1

Similarly, the Artificial Intelligence Needs Assessment Survey in Africa highlights a proliferation of community-based AI courses, hackathons and start-ups, but stresses that without coherent national strategies and investment in teacher capacity, these pockets of innovation will not transform entire systems. UNESCO+1

Here, post-secondary institutions are both beneficiaries and drivers of change: they must train the teachers who will introduce AI in schools, even as they struggle to adapt their own curricula.

4.3 A new geopolitical map of AI in African education

The future funding and governance of AI in African universities is also becoming geopolitical.

In 2025, the United Arab Emirates announced a $1 billion “AI for Development” initiative aimed at expanding AI infrastructure and services across Africa, with education listed as one key sector alongside health and climate adaptation. Reuters

Such initiatives can help fill critical gaps in infrastructure and training. They can also create new dependencies and influence, as donors shape priorities and platforms. African universities risk becoming testing grounds for other countries’ AI ambitions unless they can negotiate partnerships that align with their long-term educational and research agendas.

5. Supporters: AI as Africa’s long-awaited lever

Many African policymakers, technologists and educators see AI as a historic opportunity.

The African Union’s Continental AI Strategy frames AI as a driver of economic growth, innovation, employment and even cultural renaissance, linked to Agenda 2063 and the Sustainable Development Goals. It explicitly calls for Africa-centric, ethical and equitable AI that can help solve “some of Africa’s most pressing challenges,” from health to agriculture to education. Union Africaine+2Union Africaine+2

Supporters in higher education argue that:

• AI can extend the reach of scarce faculty by automating routine tasks and providing basic tutoring.
• It can reduce inequalities inside systems, by giving underprepared students tools to catch up.
• It can help African universities leapfrog legacy systems, adopting cloud-based, AI-enhanced platforms instead of building expensive physical infrastructure they cannot maintain.

Researchers examining generative AI in higher education from a multicultural perspective note that many students around the world already consider AI tools indispensable study partners, especially in STEM fields, and that outright bans are unrealistic. SpringerLink+1

From this optimistic view, the real risk for African post-secondary education is not using AI too much, but too little — arriving late to a transformative technology and cementing existing global inequalities.

6. Critics: data colonialism, cultural erasure and the illusion of neutrality

Critics, however, warn that AI could deepen the very inequalities it promises to solve.

A recent UNESCO article on “the cultural cost of AI in Africa’s education systems” argues that imported AI tools may erase indigenous values and local knowledge traditions, especially when models are trained mainly on Western content and languages. It calls for “culturally grounded AI” and local ownership to avoid reproducing epistemic injustice. UNESCO

The NORRAG piece on AI and unequal knowledge in the Global South highlights how biases in training data and algorithms can discriminate by race, gender, language and geography, and how the concentration of AI expertise in the North risks turning Southern users into mere data providers and consumers. Norrag Education

Other scholars speak of “data colonialism”: the extraction of behavioural and educational data from students and staff to feed proprietary models controlled elsewhere, without clear benefits returning to the communities that generate that data. Brookings+1

From this perspective, the question for African universities is not just how to regulate plagiarism or handle AI-generated essays. It is how to defend human agency, cultural diversity and academic freedom in systems increasingly mediated by opaque algorithms.

7. Four choices that will shape the future

The future of post-secondary education in the era of AI is not a fixed destiny written in code. For the Global South — especially Africa — it will be determined by a series of choices made in ministries, senates, faculties and student unions over the next decade.

7.1 From prohibition to principled use

Total bans on AI in universities are unlikely to hold; students will use the tools anyway, just more secretly and unevenly. The more realistic and ethical path is principled use:

• clear guidelines on what counts as legitimate assistance versus academic dishonesty,
• assignments redesigned to emphasize critical reflection, oral defence, and local case analysis,
• explicit teaching on how AI systems work, where they fail, and how to question them.

In this sense, AI literacy should become as fundamental as information literacy or basic statistics, especially in teacher education and professional programs.

7.2 From consumers to co-creators

African universities face a strategic choice: adopt AI tools as they are, or help design and govern them.

Becoming co-creators means:

• investing in local AI research and graduate programs,
• forming consortia to negotiate with global providers,
• building open-source, low-resource language models that reflect African languages, histories and priorities,
• insisting on data protection and fair licensing in contracts.

The African Union’s strategy and UNESCO’s frameworks both stress the need for local capacity and ethical guardrails. Union Africaine+2Institut Afrique Capacités+2 But without concrete investments and political will, those principles will remain on paper.

7.3 From narrow skills to holistic formation

AI will change the labour market in ways no one fully controls. Universities cannot simply chase today’s “hot” skills. Instead, they must double down on what is hardest to automate:

• critical thinking,
• ethical reasoning,
• capacity to work with others across cultures and disciplines,
• deep understanding of local contexts and global systems.

If AI writes code and drafts reports, post-secondary education in Africa — and everywhere — must focus on forming citizens and professionals who can ask better questions of both humans and machines.

7.4 From imitation to leadership

Finally, African universities have an opportunity to lead global debates on AI in education precisely because they sit at the intersection of scarcity and innovation.

They know, more than many northern institutions, what happens when technology arrives without infrastructure, or when reforms are imposed from outside. They understand the realities of teaching across multiple languages and traditions. They live daily with the tensions between decolonising curricula and adopting global standards.

These experiences are not obstacles to be overcome; they are sources of insight the rest of the world badly needs.

8. A future still open

In wealthy countries, discussions about AI and universities often revolve around plagiarism policies, copyright and the fate of the term paper. From the vantage point of the Global South, especially Africa, the horizon is wider and the stakes are higher.

AI could help turn overcrowded, under-resourced campuses into more responsive, inclusive learning environments — or it could harden hierarchies, centralize control and export a narrow vision of knowledge that leaves African voices on the margins.

The difference will not be made by the latest model release from California or Shenzhen, but by choices taken in Dakar, Accra, Addis Ababa, Nairobi, Johannesburg, Kinshasa and countless other cities — by rectors and ministers, lecturers and students who refuse both naïve techno-optimism and cynical resignation.

The future of post-secondary education in the era of AI is still unwritten. The Global South should not be a footnote in that story. It should be one of its authors.