Nigeria’s 2027 election may be decided by algorithms before anyone votes

Nigeria’s 2027 election campaign has already started. Just not in the way most people expect.

While political parties prepare for the polls, a quieter contest is underway inside the data pipelines, recommendation algorithms, and artificial intelligence (AI) tools shaping what Nigerian voters see, read, and believe online. Experts say this tension may matter as much as anything that happens at the ballot box.

That was the central warning at Open Data and AI Day 2026, hosted by civic technology organisation BudgIT and its innovation hub CivicHive in Lagos on Wednesday, March 11.

With presidential elections scheduled for January 16, 2027, Nigeria is entering its most consequential campaign season in a digital environment saturated with AI-generated content, micro-targeted political messaging, and deepfake videos, and experts say the country’s institutions are not yet equipped to manage any of it.

Joseph Amenaghawon, acting country director at BudgIT, opened the event by framing the urgency plainly.

“In the pre-election year of 2026, we can clearly see what AI can bring in terms of influencing the rest of us on who to vote for or who not to vote for,” he said. “The critical question now is how citizens can clearly discern what is true and what is not.”

Amenaghawon warned that the months ahead would leave many people vulnerable. “The coming months will be exciting, but they will also leave a lot of people vulnerable to misinformation and fake news across Africa,” he said.

AI is not new, but its scale is

For Gbenga Sesan, founder of the digital rights organisation Paradigm Initiative, the alarm about AI in elections is less about novelty than about acceleration.

“AI is not new,” Sesan said during his keynote. “I studied engineering in 1995, and we were already working on neural networks and machine learning. What has changed today is the scale and the speed.”

The data feeding those systems has grown enormously. Nigerians consumed 13.2 million terabytes of internet data in 2025 alone, according to the Nigerian Communications Commission (NCC) data. Every search, post, and interaction leaves a digital footprint that AI systems can analyse to build detailed portraits of voter behaviour and political preference.

“You have no idea how much content is generated online every minute,” Sesan said. “And every time you search, every time you interact online, you leave a footprint.”

The rise of data-driven campaigns

Traditional political campaigns once relied largely on rallies, broadcast media, and broad polling data to reach voters. Today, they increasingly depend on sophisticated data analytics to understand voter behaviour and tailor messages with far greater precision.

In Kenya’s 2022 elections, William Ruto’s campaign abandoned traditional ethnic bloc targeting in favour of a data-driven strategy built around economic identity. 

Rather than targeting specific tribes, the campaign identified the “Hustlers”—small traders, boda-boda riders, and unemployed youth across ethnic lines—and crafted messaging around “bottom-up economics” to appeal directly to their economic realities. 

By 2024, reports suggested the strategy had evolved further, with AI-driven sentiment analysis on platforms like TikTok and Instagram to track in real time how Gen Z—the fastest-growing voting bloc—responds to policy messages.

In South Africa’s 2024 general election, an organisation called Rivonia Circle deployed an AI chatbot named “Thoko” to answer voter questions about registration and policy. Analysts simultaneously warned that comparable tools were being used to flood social media with automated, highly targeted messages designed to exploit community anxieties around immigration and land.

“The data architecture of modern politics has changed,” Sesan said.

Campaign strategists can combine voter registration data, social media activity and consumer behaviour to build detailed voter profiles. AI can then analyse these profiles to predict preferences and deliver highly personalised political messages.

“You will get specific messages tailored to you based on your behaviour,” Sesan said.

This approach, known as micro-targeting, allows campaigns to send different messages to different groups of voters. While one voter might see messages about economic policy, another might receive content designed to appeal to cultural or religious concerns.

Such personalisation can significantly influence how political narratives spread within society.

Deepfakes are already in circulation 

Beyond targeting, AI is enabling something more disruptive: the mass production of convincing disinformation through deepfakes. 

A viral video that circulated on X this year appeared to show President Bola Tinubu responding defiantly to a war threat from Donald Trump, followed by an AI-generated version of Trump mocking his competence. 

Fact-checkers at Dubawa, a fact-checking and verification organisation, confirmed it was a deepfake, noting unnatural body movements. A separate fabricated video depicted Nnamdi Kanu, leader of the separatist group Indigenous People of Biafra (IPOB), being served food in a prison yard in Sokoto.

“People will make you say things you never said,” Sesan warned.

Modern generative AI has made such content significantly harder to detect. And the problem is not only fake videos, but the coordinated networks of automated accounts that spread them. During election campaigns, such material can spread rapidly across social media platforms, confusing voters and damaging reputations.

“We know for a fact that troll farms are activated before, during and after elections to shape public conversation,” Sesan said. “Some of the comments you see online are not even from real people. They are part of coordinated systems designed to influence conversation.”

A study published in the journal Nature found that AI chatbots tested during the 2024 US election and 2025 elections in Poland and Canada were significantly more effective at persuading undecided voters than reinforcing the views of committed supporters, suggesting that AI-driven influence may matter most precisely where elections are decided.

Nigeria may not be ready

Despite growing enthusiasm for AI across Africa, researchers at the event warned that Nigeria lacks the institutional infrastructure to manage its use in elections responsibly.

“I personally believe that Nigeria as a country is not prepared to deploy AI into its electoral processes because we do not have the foundation set,” said Temitope Asama, a researcher at Machine Learning Collective Africa. “We do not have the bricks laid for the foundation that it requires.”

Asama said responsible deployment would require independent auditing, continuous oversight, and transparent governance frameworks—structures Nigeria has not yet built.

“When an AI model fails, holding only the organisation that deployed the model accountable is simply not enough,” she said. “For a process as critical as elections, there must be continuous auditing, regulation, and testing by both government and independent bodies.”

Independent National Electoral Commission of Nigeria (INEC), Nigeria’s electoral body, has set up an internal unit to examine how AI could improve the electoral process, but the unit has not yet rolled out any programmes. A representative at the event pointed to potential applications, including facial recognition for voter accreditation and AI-assisted detection of manipulated content.

“Overall, AI presents opportunities to improve electoral integrity while helping us detect and prevent fraudulent activities,” said Tope Adenugba, who represented the commission at the event.

The real contest

For Sesan, the bigger risk is not that AI will be used against democracy, but that citizens will disengage from a process they no longer feel they can understand or influence.

“We are not passive observers,” he said. “If citizens believe only politicians understand AI and data, then they will lose their voice in the process.”

With voter turnout already declining in recent Nigerian elections, the 2027 cycle will test whether AI becomes a tool for broader democratic participation or a mechanism for shaping outcomes before most voters have made up their minds.

“When you map the scale and speed of data processing to democracy,” Sesan said, “elections can finish before they start.”