Yiaga said it would combine its Process and Results Verification for Transparency (PRVT) methodology with its Election Result Analysis Dashboard (ERAD 2.0) to assess both the electoral process and the integrity of the results.
Election Observer group, Yiaga Africa, says it will deploy an artificial intelligence-assisted system alongside its election observation methodology to independently scrutinise results from Saturday’s Osun State governorship election and detect possible discrepancies between votes cast at polling units and results uploaded to the Independent National Electoral Commission’s (INEC) results portal.
Mr Itodo explained that Yiaga would combine its Process and Results Verification for Transparency (PRVT) methodology with its Election Result Analysis Dashboard (ERAD 2.0) to assess both the electoral process and the integrity of the results.
The PRVT is a statistically based observation methodology through which Yiaga Africa independently assesses election-day processes and verifies whether official results are consistent with the votes recorded at sampled polling units.
The organisation says the methodology is designed to provide an independent, data-driven assessment of election results.
ERAD is an AI-assisted and human-verified platform used by Yiaga to audit results published on INEC’s Result Viewing Portal (IReV).
The organisation previously deployed the system to analyse electronic transmission of results, including during the 2022 Ekiti and Osun governorship elections.
Mr Itodo said the presence of results on IReV should not, on its own, be treated as proof that those results are authentic.
According to him, there are stages in the electronic results-management process that are not sufficiently visible to the public, including validation processes that take place between the polling unit and the eventual publication of results.
He said this creates the need for an independent verification mechanism that can compare what was recorded at polling units with what eventually appears on the results portal.
He explained that once results move from the polling unit, there are stages of validation that are not necessarily visible to the public.
Mr Itodo said the more important question was what would happen if results uploaded to IReV had themselves been manipulated or falsified.
“That is why we say that in this entire journey, these are not visible, and INEC is to make it visible. But that’s where the PRVT meeting comes in because it helps check some of the results that are here,” he said.
Yiaga’s position is that electronic publication does not automatically settle questions about the integrity of an election. Rather, the results uploaded by INEC must be capable of being independently tested against what observers witnessed and recorded at polling units.
This distinction is important because Yiaga’s methodology does not simply reproduce INEC’s figures. Its PRVT system uses a representative sample of polling units to generate independent estimates against which official results can be assessed.
The organisation cited its experience from the 2026 Ekiti governorship election, particularly the pattern of results uploaded to INEC’s Result Viewing Portal (IReV), as a benchmark for what it would monitor during the Osun election.
Mr Itodo also warned that the integrity of the results could not be separated from the quality of the process that produced them.
“Let me also say that for us, when you have a compromised process, it cannot produce a credible outcome,” he said.
“So if the process is severely compromised, then we won’t be able to speak to the integrity.”
The statement suggests that Yiaga’s assessment of Saturday’s election will go beyond the figures eventually declared by INEC.
The organisation will also examine how the election was conducted, including the deployment of materials, voting, counting, transmission and collation of results.
This is consistent with its broader Watching The Vote (WTV) programme, which monitors opening of polls, voting and results collation, while using PRVT to independently verify official results.
Mr Itodo said the organisation had previously withheld public validation of results where it lacked sufficient confidence in the integrity of the process.
“In fact, let’s say in one of the elections, actually in the 2023 elections, we didn’t make this public because we were not confident with the results, and we’ve done this in previous times,” he said.
He therefore indicated that Yiaga Africa would not automatically endorse the outcome of the Osun election simply because INEC declares a winner.
Instead, the organisation will examine its own evidence and determine whether the official results fall within the range supported by its independent observation data.
Mr Itodo further cited the recent Ekiti governorship election as an example of the kind of results-transmission pattern Yiaga Africa would be watching for in Osun.
“For Ekiti, this was how the results trickled. The first results came at 4:10 p.m. The next result came at 4:14 p.m. There were over 283 results that hit the portal at once,” he said.
He also pointed to a period between 6:32 p.m. and 6:37 p.m. when several results appeared on the portal in quick succession.
According to him, such patterns are important to Yiaga because unusual concentrations or sudden “dumps” of results can affect the organisation’s ability to independently analyse the flow of results.
