INEC’s Suspected AI-Generated Photos: Testing Trust, Cybersecurity, Accountability Ahead of 2027
The controversy over suspected AI-generated photographs in the Independent National Electoral Commission’s (INEC) preliminary voters’ register has prompted Nigerians to question whether modern...
The controversy over suspected AI-generated photographs in the Independent National Electoral Commission’s (INEC) preliminary voters’ register has prompted Nigerians to question whether modern technology could undermine trust in the electoral process ahead of the 2027 general elections.
Technology has become essential across sectors in the 21st century, including electoral administration, where its growing use has reshaped how elections are conducted, monitored, and assessed. From biometric voter accreditation to electronic transmission of polling-unit results, it is now a key part of Nigeria’s electoral process.
While some stakeholders expect INEC to expand its use of technology in the coming elections, emerging artificial intelligence (AI), cyberattacks, and digitally manipulated information could threaten its operations and the credibility required by the Nigerian Constitution.
The controversy over the detection of AI-generated photographs arose after INEC displayed its preliminary voters’ register at Registration Areas nationwide from October 9 to 15, 2026, as required by law. The display allowed citizens to check for inaccuracies and submit claims or objections before the register was finalised with the required information for every eligible voter, as mandated by the Electoral Act.
After the preliminary voters’ register was displayed at registration centres, Nigerians urged the commission to publish it on its official website for greater transparency and public scrutiny, as stipulated by Section 19(1) of the Electoral Act 2026. On October 26, the commission uploaded the same register to its online portal.
However, within 24 hours of the online release, stakeholders, including Nigerians and civil society groups, questioned the generic appearance and unusually smooth faces in photographs attached to numerous voter profiles. They suspected that some images were artificially generated rather than genuine photographs of registered voters.
Further scrutiny revealed that voter identification numbers shown alongside the profiles were partially masked on the website. This raised questions about the authenticity of photographs linked to voter records, how submitted information was verified and what safeguards prevented potentially fabricated images from entering the official electoral database.
Videos and screenshots flooded social media as users highlighted consecutive pages of seemingly mass-produced, AI-generated profiles and duplicate voter identification details and names of deceased individuals. Nigerian political economist Professor Pat Utomi joined Nigerians and other stakeholders in questioning the implications of suspected AI-generated images in a database used for the country’s electoral process.
In a post on X, Professor Utomi wrote: “INEC cannot afford to be careless with the credibility of Nigeria’s electoral process. The presence of AI-generated images in a register meant to support our democracy raises serious questions about verification, transparency, and accountability. Our democracy is at stake here!”
Responding to allegations that it had uploaded AI-generated images of nonexistent individuals to its official voter register and concerns about election credibility, INEC acknowledged Nigerians’ concerns. It said the register remained under review, with questionable records being examined and possible technical or procedural lapses under investigation.
INEC Blames Registration Officers But Suspected AI Photos Raise Insider Security Concerns
The controversy deepened when the electoral commission said it had traced the suspected AI-generated and synthetic images in its preliminary voters’ register to alleged misconduct by some of its registration officers who had access to the register on its website.
INEC’s Head of Information and Communication Technology, Lawrence Bayode, disclosed this during an Arise Television interview on Friday about the newly unveiled voter register. He pledged that the commission would address the growing controversy.
“What we have found out is that some of these registration officers, not only in the place that we are seeing even on social media, some of them, not all of them now, some of them across the states nationwide, have done different things to bring these irregularities onto the register.”
The ICT director said the commission had identified how some irregularities occurred but could not explain why the officers had acted as they did. He acknowledged that INEC had not used AI for voter registration and was surprised that it had been introduced into the process.
“When the ABIS was run on the system, we saw these people. We know what they did. We know how they did it. But honestly, I don’t know why. Even though we have not deployed AI on this voter registration, I can tell you how this AI-generated photo found its way onto the register. I can assure you that all these irregularities that we have seen and that the Nigerians have seen, we have removed from the register.”
This development has introduced an important discussion into the controversy, as it not only reflects on the ability of artificial intelligence to manipulate electoral information, but it also exposes the risks associated with human access, internal controls, and the management of sensitive digital systems.
This raises a critical cybersecurity question: Does the threat to electoral integrity come from technology itself or from people using it without adequate oversight and accountability? Technology depends on human choices, so even the most sophisticated electoral database can be vulnerable to abused access privileges, weak verification procedures, or failed internal monitoring.
INEC’s intention to incorporate AI and the question of preparedness
Another elephant in the room that needs to be questioned is the timing of this controversy and the INEC’s decision to begin to incorporate artificial intelligence into its operational strategy for strengthening electoral administration, particularly as far as preparedness for technology and its upside is concerned.
INEC Chairman Professor Joash Amupitan previously outlined the limits of AI and technology in the commission’s work, noting their potential to improve election logistics, voter registration and results verification.
Speaking at the AI and Democracy Forum 2026 in Abuja, he pointedly explained that AI and other emerging technologies are necessities that the electoral body would adopt for elections. However, they stressed that they have auxiliary roles to play and must be guided by human integrity, transparency, and public auditing.
“Algorithms will not and must never decide the 2027 vote.” The choice of who leads Nigeria belongs exclusively to the Nigerian people. “No critical electoral decision, result collation, voter accreditation, or candidate qualification will ever be fully automated or delegated to an autonomous system. Human accountability, statutory oversight, and judicial review will remain paramount at every stage of the electoral chain,” he emphasised.
To put these principles into practice, Amupitan announced the commission’s AI governance policy framework to guide its use of AI and address limitations that could undermine the electoral process. However, the ongoing public controversy and INEC’s efforts to catch up suggest that announcing a framework or establishing a specialised unit does not demonstrate readiness for emerging challenges and uncertainties.
Nigerians need to understand how the proposed systems will operate, who will supervise them, how their accuracy will be assessed, and what procedures will be available when automated systems make mistakes.
Technology, Democracy, and the Risk of Eroding Trust
Any strategy or tool must uphold the credibility of the electoral system, a foundation of sustained democratic governance. Past elections have shown that legitimacy depends not only on how votes are cast and counted but also on citizens’ confidence in the institutions conducting them.
Technology can strengthen the electoral process by improving access to information, reducing administrative delays, supporting voter identification, and making procedures more transparent. However, it can also introduce uncertainty when its operations lack transparency, its vulnerabilities are poorly managed, or its administrators are not held accountable.
Even with advanced biometric equipment, electronic databases and AI-powered verification, a nation may struggle to hold credible elections if its electoral institutions lack integrity, competence and transparency.
Credible elections therefore require more than functioning technology. They need institutions that can account for decisions made through technological systems, ensuring those systems strengthen democratic participation rather than add uncertainty between citizens and the institutions entrusted with their votes.
Although technology is essential to modernising electoral administration, overreliance on digital systems can create further problems. INEC is therefore expected to avoid suggesting that AI alone can resolve longstanding electoral challenges: technology cannot compensate for institutional weaknesses, inadequate transparency, or insufficient accountability.
Call for Stronger Cybersecurity Controls, Insider Accountability, and Independent Audits
Reacting to the integration of technology by the electoral commission, Anthony Dibia, a cybersecurity specialist at Appclick, called for cybersecurity controls, accountability, and the need to work with an independent body to audit and maintain the synergy between technology and INEC officials to achieve credible elections.
He further highlighted the importance of strengthening both the technological and human components of electoral security. According to him, the possibility of insider involvement makes it particularly important for INEC to review how staff members access, manage and modify sensitive voter information.
He argued that a stronger firewall alone cannot address an authorised registration officer’s deliberate submission of questionable records. Internal controls must limit each user’s actions, require independent approval for sensitive actions and reliably record every change.
Consequently, Anthony recommended that INEC adopt a comprehensive defence-in-depth cybersecurity approach, involving several independent layers of protection so that the failure of one security control does not automatically compromise the entire system.
He identified stronger staff access controls as an immediate priority. Officers who create, review and approve voter records should have clearly defined, role-based permissions, and their accounts should use phishing-resistant multi-factor authentication. Wherever possible, record creation, approval and administration should be kept separate.
This would reduce the possibility of one individual independently introducing questionable information and approving it without additional scrutiny. It would also make it easier to identify and investigate suspicious activities within the registration system.
Beyond staff access, Mr Dibia believed that INEC must also examine the security of the underlying technology supporting its electoral infrastructure. Electoral portals depend on several interconnected components, including web servers, application frameworks, programming languages, databases, APIs and third-party software components. Vulnerabilities in any of these areas can create opportunities for attackers to gain unauthorised access, manipulate information or compromise sensitive records.
“AI itself should be part of the solution, not something that INEC either blindly trusts or completely avoids. It can help analyse large volumes of registration data, identify unusual patterns, detect potential biometric duplicates, monitor security logs and flag manipulated media. However, AI systems can produce false positives and false negatives, and sophisticated attackers may deliberately attempt to evade them. Consequently, decisions affecting voter eligibility and the validity of election results must remain subject to appropriate human review, established procedures and verifiable evidence,” he concluded.
No Comment! Be the first one.