2027: Experts back AI checks on voters’ register

Ahead of the 2027 general elections, experts say Artificial Intelligence could help the Independent National Electoral Commission detect duplicate voter registrations and strengthen the integrity of the voters’ register.

Speaking to The PUNCH ahead of the forthcoming PUNCH Media Foundation webinar on AI, the experts also warned that a fake notice announcing the postponement of voting could keep people away from the polls.

The warning follows the Ekiti State incident where an AI-generated image of voters circulated on social media.

Recall that during the June 20, 2026 governorship election in the state, an image purporting to show voters queuing to sell their votes went viral.

But fact-checkers at the Centre for Democracy and Development concluded it was AI-generated.

Another image circulated as evidence of massive voter turnout in Ado-Ekiti, but an analysis by the Nigerian Fact-checkers’ Coalition showed it was AI-generated.

Last week, the INEC said its biometric system flagged 268,000 records over suspected AI-generated photographs and other irregularities.

The commission further admitted that internal and field registration officers used their authorised login credentials to introduce the AI images and other irregularities into the voters’ register.

INEC also acknowledged that some genuine photographs submitted by voters had been digitally enhanced using AI-powered phone applications.

Providing further details on the figures, INEC’s Director of Information and Communication Technology, Dr Lawrence Bayode, told The PUNCH that the system had flagged 268,000 records but declined to disclose how many had been reviewed.

Bayode stated that the Automated Biometric Identification System will flag multiple and underage registrations, among others, adding that, “Human Adjudication exercise will remove them. Those that we have seen, we are cleaning up.”

He, however, clarified that photographs enhanced with artificial intelligence would not automatically lead to the exclusion of affected voters, distinguishing between genuine photographs improved with AI applications and images generated entirely by artificial intelligence.

“Those we know it is the AI on the phone that helped enhance their pictures; we can’t exclude such people because it was the AI on the phone that did that.

“Such people will still be on the register, but those AI-generated photographs, all of these will be removed. We are already looking at them,” he said.

INEC’s ICT Director assured voters of the use of ABIS to detect underage voters and multiple voting, and urged Nigerians to verify information on the commission’s media platforms.

“The platform where you can confirm if an election is postponed or cancelled is on INEC media platforms,” Bayode said.

Bayode explained that some individuals allegedly exploited a special registration option designed for persons with disabilities who have incomplete fingers or no fingers to create questionable voter profiles.

He said the option, known as special registration, was among the loopholes allegedly exploited by those responsible for the irregularities detected by the system.

The ICT director said the commission had commenced the process of removing the affected records from the preliminary register.

Reacting to INEC’s discovery of suspected AI-generated photographs in its voters’ register, a Facebook user, Edos Osayande, wrote, “But they told Nigerians they are going to use AI for elections na. This is part of the AI.”

Amupitan defends AI

However, INEC’s Chairman, Prof Joash Amupitan (SAN), defended the commission’s use of AI in election management, warning against misinformation and insisting that the technology should not be regarded as a threat to democracy.

“AI is not an enemy of democracy,” he said at the Beyond the Ballot Summit in Abuja on November 2.

Amupitan said the technology was not a threat to democracy despite its potential misuse by malicious actors, adding that it could improve electoral administration through efficient voter registration, logistics planning, data analysis and the management of large voter databases.

“Artificial Intelligence is not an enemy of democracy,” Amupitan added, though he expressed concern over the growing accessibility of generative AI tools, which he said could be exploited to manipulate voters and spread disinformation during elections.

“However, the rapid democratisation of generative AI models has dramatically lowered the barrier to entry for malicious actors; capacities and capabilities that require state-level intelligence resources, such as high voice cloning of public figures and hyper-realistic synthesis of election officials making false announcements, are now accessible at negligible cost. These tools target specific vulnerabilities with distinct timelines,” he said.

Experts back AI

AI experts who spoke to The PUNCH said the technology could help INEC identify anomalies in the voters’ register, improve voter education, detect electoral irregularities and monitor election-related threats.

They further noted that AI tools could improve the administrative side of elections by automating biometric identification and tracking electoral materials.

They added that it could help voters understand electoral procedures and access reliable information in local languages.

AI tools such as natural language processing, the experts noted, could help monitor public feedback, conversations, voter complaints, security threats and queries from different parts of the country.

A Professor of Computer Science and Artificial Intelligence at Babcock University, Remo, Ogun State, Oludele Awodele, warned that AI could also be used to create deepfakes, fabricate audio and videos, impersonate public figures and circulate false election information.

He recommended the use of watermarking and deepfake detection tools, as well as collaboration between INEC and social media platforms to identify and remove misleading content.

“There should be a mechanism in place, like some of these detection tools that INEC should be able to use to ensure that they pull down that kind of AI-generated disinformation or misinformation,” he said.

Awodele also called for sanctions for people who create, sponsor and circulate misleading AI-generated content.

The Acting Director of the Safeguarding Centre, University of Lagos, Dr Abiola Akiyode-Afolabi, said AI could help identify sexually explicit deepfake videos targeting women candidates and also protect women and persons with disabilities.

The women’s rights advocate further identified the Violence Against Persons (Prohibition) Act 2015 as an existing legal framework that could be used to protect women from AI-facilitated violence, especially when manipulated content targets them.

She noted, however, that although the law did not specifically address technology-facilitated violence or artificial intelligence, its provisions on violence in different forms could offer some protection.

Akiyode-Afolabi said safeguards would be necessary before, during and after the elections to prevent such abuses.

Also speaking, a certified information security professional, Precious Olives, said natural language processing could be deployed to monitor public feedback, speeches, social media conversations, voters’ complaints, threats and security concerns.

He called on INEC, civil society organisations and security agencies to establish coordinated early-warning and rapid-response systems to identify and debunk false information before it spreads widely.

Olives urged technology companies and social media platforms to strengthen content moderation, identify AI-generated material and introduce measures to limit the spread of misleading content.

PUNCH webinar

The PUNCH webinar, themed “AI, Misinformation and Nigeria’s 2027 Elections: Safeguarding the Truth, Protecting the Vote,” will examine how Nigeria can harness the benefits of AI while preventing its misuse to spread false information, manipulate public opinion and weaken trust in the electoral process.

The webinar will be held on Zoom on Wednesday, November 11, from 11 am to 1 pm WAT and will bring together representatives from the media, electoral institutions, security agencies, civil society and the technology sector.

The discussion will examine the risks posed by AI-generated misinformation, the responsibilities of different stakeholders and possible measures for protecting electoral integrity and public trust.

The webinar will feature the Inspector-General of Police, Olatunji Disu; INEC National Commissioner and Chairman of its Information and Voter Education Committee, Mohammed Haruna; Director of Programmes at YIAGA Africa, Cynthia Mbamalu; Director of Projects at Cable Newspaper Journalism Foundation, Mayowa Tijani; and Founder of FactsMatterNG, Hannah Ajakaiye.

It will be moderated by The PUNCH Metro Editor, Joel Nwokeoma.

Here’s the registration link: https://tinyurl.com/2027-Ai-Election-Webinar