Orodata, a Nigerian data and civic technology organisation, has highlighted the risks of artificial intelligence (AI)-generated political messaging, following a reported influence attempt in Kenya ahead of its 2027 general election.
Orodata raised the concern in a publication titled “Manufactured Grassroots: Lessons for Civil Society from a Reported AI Influence Attempt in Kenya.”
The report was released officially to the media on Tuesday in Lagos.
The publication draws on a September 2026 threat intelligence report by Anthropic, which documented the use of AI to generate social media posts designed to appear as spontaneous grassroots political commentary.
The Kenyan case involved a pro-government operator who used AI to prepare political messages ahead of the country’s 2027 general election.
The development raises concerns about how AI could be used to manufacture the appearance of public support for political positions, particularly as Nigeria approaches its 2027 general elections.
The reported operation involved generating batches of social media posts that appeared to represent independent public opinion rather than coordinated messaging.
According to Anthropic, the content included messages praising a government official over a proposed electricity tariff increase and narratives suggesting that opposition politicians were divided.
The operation demonstrated how AI could help produce political messages at scale while making coordinated communication appear to be the views of ordinary citizens.
Anthropic said it identified and disrupted the activity. Its assessment found that the campaign remained within a network of inauthentic accounts and did not achieve significant engagement beyond that network.
The company also reported no evidence of government involvement in the operation.
The report has implications for Nigeria.
For Nigeria’s civil society organisations, the case highlights the importance of distinguishing genuine public participation from organised attempts to create misleading impressions of popular opinion.”
As political communication increasingly takes place online, AI-generated content could complicate efforts to establish whether widely circulated political messages reflect authentic citizen views.
The case also points to the importance of scrutinising digital political campaigns without confusing legitimate political expression with coordinated manipulation
.
“For civil society groups, journalists and other stakeholders, the challenge is to recognise potentially misleading online activity while protecting citizens’ ability to express political opinions freely.
The report said that the Kenyan experience provides a relevant case for Nigeria as stakeholders consider the implications of AI-assisted political communication for public debate and democratic participation ahead of the 2027 elections.
