The Independent National Electoral Commission (INEC) has warned that artificial intelligence (AI) could be exploited to fabricate election results, impersonate electoral officials and incite violence ahead of the 2027 general elections.
INEC Chairman, Prof. Joash Amupitan, stated this on Monday at the National Summit on Electoral Information Integrity in Abuja.
Amupitan called for coordinated response by the commission, security agencies, media organisations, technology companies and fact-checking groups to tackle the growing threat of AI-generated misinformation.
The summit, themed “Beyond the Ballot: AI, FIMI and the Integrity of Nigeria’s 2027 Elections,” was convened by the International Press Centre (IPC), the Centre for Media and Society (CEMESO) and DAI under the European Union Support to Democratic Governance in Nigeria (EU-SDGN) Programme.
Amupitan said the growing accessibility of generative AI tools has made it possible for individuals and organised groups to produce convincing fake audio, videos and other content capable of misleading voters and undermining confidence in election administration.
He warned that fabricated results could circulate on social media before INEC makes official announcements and potentially trigger public unrest and violence.
According to him, the threat to elections now extends beyond the physical security of ballot papers, polling units and electoral officials to the digital information space before authorities have an opportunity to respond.
“Today, elections can be undermined long before the first ballot is cast,” he said.
The INEC chairman identified synthetic media, automated misinformation and disinformation, deepfakes, digital micro-targeting and Foreign Information Manipulation and Interference (FIMI) as emerging threats that could compromise the credibility of the 2027 elections.
He cited the possibility of AI-generated audio impersonating senior political figures to make false claims about vote-buying or electoral manipulation, as well as fabricated videos of electoral officials announcing decisions they never made.
Amupitan also warned that manipulated audio and video could be deliberately released during result collation and announcement to create confusion and provoke reactions before INEC communicates verified information.
He said the increasing sophistication of AI tools had lowered the cost of carrying out such operations. He, however, noted that the commission had established an AI unit to anticipate the risks associated with emerging technologies and develop measures to protect its electoral processes.
The INEC boss said AI was already embedded in many technologies used by Nigerians, including mobile phones and electoral systems, making it impossible for the commission to ignore its growing influence.
He noted that AI was not inherently harmful, noting that it could assist election management bodies with voter register analysis, logistics planning, data processing and administrative efficiency.
He said the danger depends on how the technology was deployed.
Amupitan further recalled a recent engagement in Canada involving technology companies such as Google, Microsoft and Meta, where participants discussed the difficulty of containing the growing threat of AI-generated harmful content.
He said even technology companies were grappling with the challenge because new forms of manipulation continued to emerge and require constant updates to detection systems.
He therefore called for the deployment of more advanced tools to detect and counter manipulated content.
Speaking further, Amupitan said the commission has entered into a memorandum of understanding with TikTok, and is seeking improved cooperation with Facebook, Instagram, WhatsApp, YouTube, Telegram and X.
INEC has also introduced ChatGPT to support its efforts and is working with development partners, including the European Union, to establish a strategic communication crisis network.
According to Amupitan, relying solely on fact-checking after false information had gone viral would be inadequate, as manipulated content could spread widely within 30 minutes, making it difficult to reverse the damage.
However, Amupitan cautioned against using the threat of AI-generated disinformation to justify censorship, restrictions or the suppression of political dissent.
He said measures to combat hate speech and synthetic electoral fraud must respect fundamental rights, remain transparent and proportionate, and be subject to legal accountability.
“Measures aimed at curbing hate and synthetic electoral fraud must remain strictly rights-respecting, transparent, proportionate and subject to legal accountability,” he said.
“The integrity of the 2027 ballot depends not only on the physical security of the polling unit, but on the authenticity of the information environment in which the voters make their decision,”
