Trending 
An artificial intelligence (AI)-driven influence operation has targeted former President Uhuru Kenyatta and former Deputy President Rigathi Gachagua before Kenya’s 2027 election.
The operation, identified as GTG-54004, allegedly used Anthropic’s Claude AI model to produce large volumes of political posts designed to appear like genuine grassroots commentary.
According to a threat intelligence report by global AI company Anthropic, the campaign also promoted Energy Cabinet Secretary Opiyo Wandayi. The operation allegedly pushed narratives portraying Kenya’s United Opposition as divided and close to collapse, targeting Gachagua and Kenyatta.
“The actor used Claude across several sessions to generate batches of exactly 50 tweets, explicitly instructing the model to make posts look like spontaneous grassroots commentary rather than a coordinated campaign,” the report said.
Anthropic said the network also promoted stories claiming the United Opposition coalition was breaking apart before the 2027 general election.
“Additionally, the network pushed stories claiming that Kenya’s United Opposition coalition was breaking apart before the 2027 general election, directly targeting politicians Rigathi Gachagua and former president Uhuru Kenyatta,” Anthropic said.
The campaign also promoted favourable messages about Wandayi, particularly his role in halting a planned electricity tariff increase. It used hashtags including #PowerReliefKE and #PoweringTheNewKenya, with posts presented as independent comments from ordinary social media users.
Anthropic said the operator supplied Claude with topics, talking points and hashtags before asking the AI to make the posts appear more natural and human. The same workflow was allegedly used for commercial marketing campaigns involving Kenyan retail brands.
“Separately, the actor ran the identical AI workflow for Kenyan retail brands under a marketing persona, ‘SHANKI’/’Elkins Marketer,” the report said.
Anthropic said the campaign was designed to create the appearance of organic online engagement rather than simply produce political content.
Despite the volume of AI-generated posts, however, Anthropic assessed that the operation had little impact beyond the fake accounts involved.
“We evaluated the impact of the network using the Breakout Scale and classified it as Category One. The activity was completely isolated within the network of fake accounts and local influences on a single platform, failing to reach or influence any real people,” the report said.
The assessment placed the operation at the lowest level of Anthropic’s impact scale, indicating that it failed to reach wider public discourse or significantly influence genuine users. Anthropic also found no evidence linking the operation directly to the Kenyan government or showing that it was government-funded.
The company instead classified it as a domestic Kenyan influence operation.
The network was uncovered after threat intelligence shared by OpenAI indicated suspected repeat misuse of AI systems. Anthropic subsequently banned the accounts involved and introduced additional behavioural detections to identify similar activity.
The findings highlight growing concerns about generative AI being used to mass-produce political messages and make coordinated campaigns appear organic.
