Artificial intelligence is changing how we work, learn, play, think and relate to one another. It is already transforming medicine, education, scientific discovery and economic productivity. Used well, AI could expand human capability on a scale few technologies in history have achieved.

But the same technology can also deepen inequality, industrialise misinformation, amplify political division and make it increasingly difficult to distinguish authentic human communication from synthetic persuasion. They can help citizens understand their world or manipulate what citizens believe about it. The defining question, therefore, is what kind of societies will AI help us build?

For Nigeria, that question is becoming urgent. Presidential and National Assembly campaigns officially begin on 19 August. Nigerians will vote on 16 January 2027, followed by governorship and State Assembly elections on 6 February. But much of the political contest will take place long before Nigerians enter polling stations. It will unfold on X, Facebook, TikTok, Instagram, YouTube and WhatsApp; through viral videos, memes, influencers, anonymous accounts and, increasingly, AI-generated content.

Democracy has vital signs

We tend to measure democracy through institutions: elections, constitutions, courts, political parties and legislatures. These matter enormously. But democracy also lives or dies in the quality of public discourse or the everyday encounters between citizens.

| | |
|---|

Do we still recognise the humanity of people with whom we disagree? Can political opponents argue without treating one another as enemies? Do citizens believe democratic institutions can respond to them? Do people feel capable of acting together to improve their communities? And what happens when distrust, humiliation, ethnic hostility or political hopelessness become dominant features of the public conversation?

These questions led us at the Global Centre for Rehumanising Democracy to develop the Democracy Discourse Index (DDI). Built with a growing consortium of universities, the DDI uses a combination of human research and artificial intelligence to examine the democratic quality of public discourse. We see it as taking the vital signs of the democratic public sphere. The system analyses large volumes of publicly available discourse against indicators organised around four empathy, civility, trust language and democratic agency.

This matters because dangerous political environments rarely emerge overnight. Before communities become enemies, people learn to describe one another as enemies and before citizens abandon democratic institutions, distrust and hopelessness become normalised. We have seen this pattern consistently in our studies in Rwanda, the Democratic Republic of Congo and Eastern Europe. If we can detect those shifts early enough, democratic societies have a better chance of responding.

There is another dimension to the DDI that we believe is particularly important for Nigeria. One of the most consequential questions facing every society is how we prepare young people to live responsibly in an AI-saturated world. Youth must have the knowledge, digital skills and ethical framework to use these technologies to better their societies. Education must go beyond digital literacy. Students should be trained on how to use AI to evaluate information and to engage with and improve their communities and countries.

Teaching students merely how to prompt ChatGPT will not be enough. They need critical thinking to understand bias, misinformation and algorithmic amplification and an ethical framework for deciding what technology ought to be used for.

For the past year, we have been conducting an unusual experiment with university students and faculty. Students and academics in Pakistan, Albania, Greece, Bulgaria, North Macedonia and the United States have been studying public discourse in their own societies. We are expanding to Ecuador, Lebanon, the United Kingdom, Ukraine and Nigeria.

Students do not simply consume AI. They help build and interrogate it. They examine real public conversations and learn to identify empathy, civility, trust, democratic agency and other characteristics of political communication. Their human annotations are then used to train AI systems capable of conducting similar analysis at much greater scale. This reverses an increasingly common relationship between young people and artificial intelligence.

Instead of technology merely learning how to capture their attention, young people are teaching technology how to recognise the conditions necessary for healthier human societies. That is the kind of AI education Africa should be pioneering.

Nigeria needs this capacity before 2027

The implications for the coming election are substantial. Imagine being able to detect that dehumanising rhetoric against an ethnic or religious community is accelerating in a particular part of the country before it spills into offline violence; or journalists being able to distinguish an isolated offensive post from a rapidly expanding pattern of coordinated hostility.

And imagine being able to identify communities where Nigerians are responding to political disagreement with empathy, restraint and civic solidarity and learning from them. The DDI is designed to help make those patterns visible.

Fact-checking remains essential, but identifying whether a claim is true or false addresses only part of the problem. A technically accurate statement can still be used to dehumanise. A political conversation containing no demonstrably false information can nevertheless become dangerously hostile. And societies can experience profound democratic deterioration through cynicism, distrust and political helplessness without a single viral “fake news” story causing it. We need to understand the condition of the conversation itself.

Can Africa help build artificial intelligence that strengthens human dignity, democratic agency and social cohesion?

Nigeria should be at the centre of that conversation. Its extraordinary linguistic, religious, ethnic and cultural diversity makes democratic coexistence difficult, but it also makes Nigeria one of the world’s most important laboratories for understanding how human beings negotiate difference. That knowledge and experience has global value.

On Monday, students and faculty participating in the Democracy Discourse Index consortium will meet in a global virtual forum to present what they have learned from applying this approach in different societies. The event is free and open to participants interested in exploring how universities can use AI not simply to improve productivity, but to strengthen democratic citizenship and build better futures for their societies.

As Nigeria enters its first general election of the generative-AI age, we should aim to build the intellectual, technological , ethical and civic capacity to recognise and strengthen what can hold our societies together.

Dr. Margee Ensign is Board Chair of the Global Centre for Rehumanising Democracy (GCRD). She has more than 15 years of experience as a university president across the United States, Eastern Europe and Africa, including serving as President of the American University of Nigeria.

Dr. Jacob Udo-Udo Jacob is the Founding Executive Director of GCRD. A scholar of communication and democratic governance, he has held several academic leadership positions, including serving as Dean of Graduate School and Research at the American University of Nigeria.