2027: INEC Explains How AI Will Flag Election Result Discrepancies
The Independent National Electoral Commission (INEC) is using artificial intelligence to help detect discrepancies in election result records ahead of the 2027 general elections.
The commission’s Director of Information and Communication Technology, Dr Lawrence Bayode, explained the proposed checking process on Thursday, October 22, at the launch of the AI for Elections Programme and Hackathon organised by Rise Networks in Abuja.
Bayode said the system would compare figures entered electronically with those recorded on paper result sheets and their scanned copies. Where the figures do not match, the discrepancy must be resolved before the result can be uploaded.
“The system checks whether the figures entered electronically match those on the paper result sheets and scanned copies before transmission is allowed. If discrepancies exist, they must be resolved before the results can be uploaded,” he said.
The process is intended to help electoral officials identify possible errors during result management and ensure that figures published on INEC’s portal correspond with the records from polling units.
Bayode said the commission was exploring practical applications of AI while putting safeguards in place to protect electoral security, operational stability and public confidence.
The development comes as INEC prepares for the 2027 elections amid continuing concerns about the accuracy, transmission and public verification of election results.
In September, INEC Chairman Joash Amupitan disclosed that the commission had begun deploying AI tools to help identify discrepancies in election result records. He said the technology would support electoral officials rather than replace human decision-making.
The commission has also warned about the risks posed by AI-generated misinformation, including fabricated videos, cloned voices and manipulated documents that could be circulated as genuine election information.
These risks present a challenge for election administrators and journalists, who must distinguish authentic electoral records and announcements from misleading or fabricated material.
At Thursday’s event, Meta’s Head of Public Policy for Anglophone West Africa, Sade Dada, said advances in AI were making it easier to create misleading content that could be difficult for the public to distinguish from genuine material.
She also pointed to possible benefits of the technology, including content verification and tools that could help voters access reliable electoral information in languages they understand.
The AI for Elections initiative, organised by Rise Networks, is designed to bring technology developers, researchers, electoral experts and civic groups together to develop practical solutions to challenges facing Nigeria’s electoral process.
Its areas of focus include access to credible electoral information, election data analysis, multilingual civic technology, cybersecurity and election observation.
Although AI-assisted checks could help identify inconsistencies in result records, their effectiveness will depend on how accurately they interpret source documents, how discrepancies are resolved and whether the process remains open to appropriate human and legal scrutiny.
INEC has not, in the material examined for this report, provided a detailed public explanation of the testing status, error-handling procedure or independent audit arrangements for the specific checking process described by Bayode.
The commission’s use of AI is therefore an important development in its preparations for 2027, but the technology’s role and safeguards will need to be clearly communicated to electoral stakeholders and the public.
