The Arusha mandate: Why Nigeria must lead Africa’s AI examination revolution

By Abdullahi O Haruna

Every age has its defining infrastructure. The nineteenth century built railways. The twentieth erected highways, power grids and telecommunications networks. The twenty-first century is constructing something less visible but no less consequential: systems for measuring human capability.

This may sound abstract. It is not. Societies that most accurately identify talent, certify competence, and align education with economic reality tend to outperform those that do not. Nations rise not merely because they educate people, but because they know how to distinguish excellence from mediocrity, skill from certification, and knowledge from mere familiarity.

In an era increasingly shaped by Artificial Intelligence, that distinction is becoming more important than ever. This is why a recent event in Arusha, Tanzania, deserves more attention than it has received.

The occasion was the 42nd Annual Conference of the Association for Educational Assessment in Africa (AEAA), an organisation whose name alone is enough to deter all but the most determined policy enthusiast. The gathering brought together examination administrators, educational researchers, policymakers and assessment specialists from across the continent. Their task was to discuss what, at first glance, appears to be a narrow technical subject: educational assessment in the age of Artificial Intelligence.

Yet beneath the conference theme lay a question that will help determine Africa’s economic future. How should a continent of nearly 1.5 billion people evaluate learning when machines are increasingly capable of performing many of the intellectual tasks that schools have traditionally rewarded?

The answer matters because education is changing. More importantly, work is changing. And if work changes, examinations must change too. At the Arusha conference, Africa appeared to choose who should help lead that transition.

The election of Professor Dantani Ibrahim Wushishi, Registrar and Chief Executive of Nigeria’s National Examinations Council (NECO), as Vice President of the AEAA, alongside the decision to grant Abuja hosting rights for the association’s next conference in 2027, may have seemed routine. In reality, it was a signal.

The continent was not merely selecting a conference host or filling an executive position. It was identifying a country it believes can help shape the future of educational assessment in Africa. That country is Nigeria.

Whether Nigeria fully appreciates the significance of this responsibility remains to be seen.

Educational assessment is rarely celebrated. Politicians cut ribbons at schools, not examination centres. Governments announce new universities, not new marking systems. Public debates focus far more on access to education than on how learning is measured. Yet assessment sits at the centre of every educational system.

A nation can build thousands of classrooms and recruit armies of teachers. It can spend billions on educational infrastructure. But if it cannot accurately assess what students know and what they can do, much of that investment risks becoming an expensive exercise in uncertainty.

Examinations are the quality-control mechanism of education. Without them, policymakers are effectively flying blind. For much of the past century, examination systems around the world operated on assumptions inherited from the industrial age. Students memorised information.

They reproduced it under controlled conditions. Success depended largely on recall, repetition and conformity. The model reflected the needs of its time. Factories required discipline. Bureaucracies required standardisation.

Industrial economies rewarded routine cognitive skills. Schools responded accordingly.

The problem is that the economy those examinations were designed for is disappearing.

Artificial Intelligence is accelerating that process. A growing number of tasks that once required human expertise can now be performed by algorithms. Data analysis, language translation, information retrieval, content generation and even elements of legal research are increasingly automated.

Machines are becoming better at remembering things. This means humans must become better at doing things machines cannot easily replicate. Creativity. Judgement. Adaptability. Complex problem-solving. Ethical reasoning. These qualities are becoming the new economic premium. Yet traditional examinations often struggle to measure them.

Many education systems continue to test students for a world that is already fading. The danger is obvious. Countries may find themselves producing graduates whose qualifications certify competence in skills that are declining in economic value while failing to identify the abilities that matter most in the emerging economy. Educational assessment therefore stands at a crossroads.

It can either evolve alongside technological change or become increasingly disconnected from reality. This is where Artificial Intelligence presents both an opportunity and a challenge.

Much of the public discussion surrounding AI in education has focused on students using technology to complete assignments. Teachers worry about essays written by chatbots. Universities debate plagiarism. Governments discuss regulation.

These concerns dominate headlines. Yet they may not be the most significant educational consequence of AI. The larger transformation may occur within assessment itself.

Artificial Intelligence has the potential to redesign examinations from the ground up. Consider adaptive testing. Traditional examinations operate on a simple principle: everyone answers the same questions. The system is designed for administrative convenience. But it is not necessarily the most accurate way to measure ability. Imagine two candidates sitting the same mathematics examination. One possesses exceptional competence. The other struggles with fundamental concepts. Both receive identical questions. Both produce scores. Yet the examination may reveal surprisingly little about either candidate’s actual capabilities. The stronger student is insufficiently challenged. The weaker student is overwhelmed. Artificial Intelligence offers a more sophisticated approach.

An adaptive examination can adjust itself in real time. As a candidate answers questions, the system analyses performance and modifies subsequent questions accordingly. Difficulty rises when competence is demonstrated. It falls when difficulties emerge. The result is a more precise picture of knowledge and ability. Instead of crudely categorising students as pass or fail, assessment becomes a nuanced process of measuring competence.

For countries seeking to develop highly skilled workforces, such precision could prove transformative. The implications extend far beyond individual candidates. Governments have long struggled with a peculiar problem.

Education systems generate enormous quantities of information, yet policymakers often possess remarkably little insight into what that information actually means. Every examination season produces mountains of data. Most of it serves one purpose. Determining who passed.

The remainder is largely ignored. Artificial Intelligence changes that equation. Modern analytical systems can identify patterns invisible to human observers. They can reveal regional disparities in educational performance. They can detect weaknesses in specific subjects. They can identify schools that need intervention and programmes that produce measurable results.

Examinations cease to be administrative exercises. They become intelligence-gathering systems. Ministries gain evidence rather than assumptions. Policy becomes more precise. Resources become better targeted. In countries where educational budgets are constrained and development needs are immense, this matters enormously.

Nigeria understands the scale of this challenge better than most. Every year, millions of Nigerian students sit examinations administered by various national bodies. The logistical complexity is staggering.

A weakness affecting a small country might inconvenience thousands. A weakness affecting Nigeria can affect millions. This reality helps explain why examination administration in Nigeria has become more than routine bureaucracy. At its best, it is an exercise in managing complexity at continental scale. This is one reason the developments within NECO over recent years have attracted attention.

Since Professor Wushishi assumed leadership in 2021, the organisation has pursued reforms aimed at modernisation, digitalisation and institutional strengthening. Some of these changes have been technical. Others have been operational.

Collectively, they have signalled an ambition to position NECO not merely as a domestic examination body but as an institution capable of contributing to broader continental conversations.

The council’s growing international footprint is particularly notable.

Conducting examinations beyond Nigeria’s borders requires confidence, credibility and administrative sophistication. It reflects an institution comfortable operating within increasingly interconnected educational environments. The support Nigeria received in Arusha did not emerge solely from diplomatic goodwill. It reflected recognition of these developments. Yet leadership brings obligations.

And nowhere are those obligations more apparent than in the struggle against examination malpractice.

Cheating is one of the oldest challenges in education.

Wherever examinations exist, incentives to manipulate outcomes arise. Africa is no exception. Question paper leaks, impersonation schemes, collusion networks and organised fraud have periodically undermined public confidence in educational credentials.

The consequences extend beyond individual examinations. When trust in assessment declines, so does trust in qualifications. Employers become sceptical.

Universities become cautious. Meritocracy suffers. Artificial Intelligence offers new tools in this battle. Facial recognition systems can verify identity. Behavioural analytics can identify unusual examination activity. Machine-learning algorithms can detect suspicious answer patterns across thousands of scripts within minutes.

Tasks that once required weeks of manual investigation can increasingly be performed automatically. Technology cannot eliminate dishonesty. Human beings remain infinitely inventive when pursuing unfair advantage. But AI can significantly alter the balance. It can increase the probability of detection and reduce opportunities for manipulation.

For educational systems seeking credibility, this capability is invaluable. Yet the greatest challenge confronting Africa’s assessment future may not be technological at all. It may be social. Artificial Intelligence promises efficiency. It does not automatically guarantee fairness. Nigeria illustrates the dilemma.

The country possesses one of Africa’s most vibrant technology ecosystems. Digital entrepreneurship flourishes. Mobile technology is widespread. Innovation hubs attract global attention. At the same time, vast disparities persist. Reliable electricity remains uneven. Internet access varies dramatically. Digital literacy is inconsistent. The risk is clear. If technological assessment advances faster than technological inclusion, existing inequalities may deepen. Students in urban centres could gain advantages unavailable to candidates in rural communities.

Educational opportunity might become increasingly tied to infrastructure rather than ability. This would represent a profound failure. The objective of educational reform is to expand opportunity, not redistribute it according to technological access.

Policymakers therefore face a delicate balancing act. They must embrace innovation without sacrificing equity. They must modernise assessment while preserving inclusiveness. They must ensure that technological sophistication strengthens public trust rather than undermines it. This requires more than software. It requires governance. Too often, discussions about Artificial Intelligence descend into a form of technological determinism—the belief that innovation itself will solve institutional problems.

History suggests otherwise. Technology amplifies capacity. It rarely substitutes for competence. Countries with strong institutions tend to benefit disproportionately from technological change because they possess systems capable of deploying innovation effectively.

Countries with weak institutions often discover that technology merely reproduces existing dysfunctions in more sophisticated forms. The lesson is straightforward. Artificial Intelligence is not a strategy. It is a tool. The strategy must come from leadership. This is why Abuja’s selection as host city for the 2027 AEAA conference carries symbolic importance.

By the time delegates gather in Nigeria’s capital, the educational landscape will have evolved considerably. AI systems will be more powerful. Assessment technologies will be more sophisticated. The ethical questions surrounding data privacy, algorithmic bias and automated decision-making will become more pressing. The conference will not merely discuss possibilities. It will evaluate outcomes.

Delegates will arrive seeking evidence that developing countries can successfully integrate advanced technologies into educational systems without compromising fairness or accessibility. They will look for practical models.

Nigeria will inevitably be under scrutiny. That scrutiny should be welcomed. For decades, Africa has often occupied the position of observer in global technological transformations. Innovations emerged elsewhere. African countries adapted as best they could.

Artificial Intelligence offers a rare opportunity to alter that dynamic. Educational assessment remains a field that is rapidly evolving. Standards are not yet fixed. Best practices are still emerging. This creates space for leadership.

The continent need not wait for solutions developed elsewhere. It can help shape them. Nigeria’s size, institutional experience and growing influence place it in a unique position to do precisely that. The events in Arusha therefore represented more than a diplomatic success.

They represented a continental expectation. Africa has effectively assigned Nigeria a role in one of the defining educational transformations of the century. The assignment comes with no guarantees. Opportunities rarely do. History is littered with countries that possessed favourable conditions yet failed to translate potential into achievement.

Leadership is ultimately measured by results. Between now and Abuja 2027, Nigeria will have the opportunity to demonstrate what intelligent, inclusive and technologically sophisticated assessment can look like in an African context.

Success would resonate far beyond examination halls. It would influence workforce development, economic competitiveness and institutional trust across the continent. Failure would be equally instructive, though considerably less desirable. Either way, the stakes are clear.

The future of African education will not be determined solely by what students learn. It will also be determined by how societies choose to measure that learning. And for the moment, at least, Nigeria holds the pen. The mandate has been granted. The continent is watching. The examination has already begun.

…Haruna writes from Abuja