Special Report
By Chukwudi George Ozalla
The Transcorp Hilton Hotel, Abuja, came alive on October 14 and 15 as the beautifully lit blue hall filled with many enthusiastic participants for the AI and Democracy Forum 2026. Representatives of electoral institutions, civil society organisations, technology companies, the media, academia and development partners assembled to examine the growing influence of artificial intelligence on elections and the challenges it could pose to Nigeria’s 2027 general elections.
The blue lighting, lively atmosphere and exhibition booths showcasing the work of different organisations added colour to the two-day event. Participants exchanged ideas, asked questions and examined practical ways to address the growing threat of digitally manipulated information to elections and public confidence in democratic institutions.
Organised by YIAGA Africa in partnership with the Independent National Electoral Commission (INEC) and other stakeholders, the forum was held under the theme, “Safeguarding Nigeria’s 2027 General Election in the Age of Artificial Intelligence.”
The programme featured five plenary sessions, seven practical masterclasses, a town hall with technology companies and high-level cybersecurity workshops for INEC national commissioners. The sessions covered AI in election administration, deepfake detection, election reporting, cybersecurity, civil society oversight and the accountability of technology platforms.
The forum followed a validation workshop on the ECOWAS Guidelines for Artificial Intelligence in Election Administration, jointly hosted by the ECOWAS Network of Electoral Commissions and YIAGA Africa. Its focus was timely as political campaigns, voter engagement and public debate increasingly depend on digital platforms where false claims, fabricated videos and AI-generated content can spread rapidly.
Speaking at the forum, INEC Chairman, Professor Joash Amupitan, declared that artificial intelligence would not be allowed to determine the outcome of Nigeria’s 2027 general election.
“Algorithms will not, and must never, decide the 2027 vote,” he said.
Amupitan explained that artificial intelligence could influence the information environment and assist with electoral logistics, but insisted that the sovereign will of voters must remain the sole determinant of electoral outcomes.
He identified possible uses of AI in route optimisation and logistics, biometric database hygiene, multilingual voter engagement, cybersecurity and assistance with result audits. Such applications, he said, could help electoral officials improve efficiency and address operational challenges, provided they were introduced with proper safeguards.
However, he maintained that critical electoral decisions, including result collation, voter accreditation and the qualification of candidates, would not be fully automated. He stressed that human accountability, statutory oversight and judicial review must remain central to the electoral process, regardless of technological advances.
“Our task, therefore, at this forum and within the commission, is to ensure that AI remains a servant of democracy, not its master,” the INEC chairman stated.
His remarks placed responsibility on electoral authorities to ensure that technological innovation does not weaken public confidence or replace the legal procedures governing elections. As the 2027 polls approach, the challenge will be to make useful technologies available without creating room for unaccountable systems or leaving voters uncertain about how electoral decisions are reached.
Executive Director of YIAGA Africa, Samson Itodo, said technology alone could not guarantee credible elections without accurate data, transparent methodologies, systems that could be independently audited and the personal integrity of those managing the process.
He drew attention to concerns among Nigerians about the credibility of elections and whether their votes would count in 2027. He noted that public confidence required more than the deployment of digital tools, particularly in view of past experiences with electoral technology.
“It is one thing to deploy technology; it is another thing for the personal integrity of those managing those technologies to deliver,” Itodo said.
He warned against the possible use of artificial intelligence-powered platforms to aggregate election results. While civil society organisations could independently collate figures for monitoring and accountability, he explained that INEC remained the only institution legally empowered to declare official election results.
The warning raised an important question about the use of AI in election monitoring: how can independent organisations use technology to strengthen transparency without confusing the public over which results are official? Clear communication, verifiable data and respect for INEC’s statutory responsibility are essential to addressing that concern.
The dangers of deepfakes and synthetic media also received attention. These technologies can be used to produce convincing videos, images and audio recordings that falsely portray candidates, public officials or electoral officers as saying or doing things that never happened. If circulated widely during a campaign, such material could mislead voters before the affected individuals or institutions have an opportunity to respond.
A session on deepfakes and synthetic media featured David Ajikobi, Nigeria Editor at Africa Check; Silas Jonathan, Digital Investigations Manager at the Centre for Journalism Innovation and Development; and Professor Suraj Olunifesi of the UNESCO University Network on Media and Information Literacy.
The session examined the difficulties of identifying manipulated content and the need for journalists, fact-checkers and citizens to verify suspicious material before sharing it. False claims can travel quickly through social media and messaging platforms, while corrections may reach fewer people than the original material.
The forum also considered the responsibilities of institutions and technology platforms in responding to these threats. Discussions on electoral safeguards and accountability featured former INEC National Commissioner, Professor Okechukwu Ibeanu; Dr Lisa Poggiali of the International Foundation for Electoral Systems (IFES); Cynthia Mbamalu of YIAGA Africa; and Sani Suleiman of Paradigm Initiative. Geoffrey Uzono of Channels Television moderated the session.
Participants examined legal frameworks, institutional responsibilities, risk assessment and accountability when artificial intelligence is used in electoral processes. These questions are important because responsibility cannot be left unclear when an automated system produces misleading information, exposes sensitive data or affects public understanding of an election.
Officials from technology companies, including Microsoft and Google, also took time to explain technology-related issues and answer participants’ questions. Their engagement gave attendees an opportunity to seek clarification on the capabilities and limitations of digital tools and the measures available to address the misuse of AI-generated content.
The practical sessions were another major part of the programme. The seven masterclasses covered AI in election administration, civil society oversight of generative AI, election reporting in the age of AI, cybersecurity and early warning, deepfake detection, election governance and the accountability of technology platforms. The training tracks were designed to help participants understand how to identify risks and apply appropriate safeguards within their respective fields.
The exhibition booths gave participating organisations room to showcase their work and demonstrate technology-related solutions. The displays complemented the discussions by allowing attendees to engage with organisations working on electoral integrity, digital information, election observation and voter engagement.
The forum brought together a range of partners, including the Centre for Journalism Innovation and Development, Dataphyte, the ECOWAS Network of Electoral Commissions, the International Foundation for Electoral Systems, GIZ, Microsoft and other national and international organisations.
The AI in Elections Sandbox Challenge featured young innovators pitching technology-based ideas for addressing electoral challenges. Election GPT by Pawavox was subsequently announced as the Judges’ Choice Winner in the Election Observation and Data category. The solution combines voice and video capture with AI-powered analysis, allowing observers to submit geotagged, time-stamped and cryptographically signed reports. The team was named to receive a ₦13 million prize, alongside technical support to develop and improve the solution ahead of the 2027 elections.
The forum’s central message was that artificial intelligence could assist election administrators, journalists and observers, but its use must be accompanied by safeguards, accountability and respect for the law. With the 2027 general elections approaching, the task for INEC, technology companies, political actors, the media and civil society will be to ensure that these tools help voters make informed choices rather than make it harder for them to establish what is true.
