Anthropic: Is Russia Using AI for Disinformation in CAR?
A report from US-based company Anthropic alleges its AI model, Claude, was used in pro-Russian disinformation campaigns in the Central African Republic (CAR). The report, titled "Detecting and Countering Misuse of AI," detailed how the AI was purportedly employed to generate propaganda and support surveillance against opposition figures.
The Anthropic report, published in September, claims that Russian agents in CAR deliberately used the Claude AI assistant. Its primary reported use was producing content for Radio Lengo Songo, a station that the international investigative collective All Eyes on Wagner states is funded by Russia. This content reportedly promoted pro-Russian narratives, portraying the Central African government and the Africa Corps positively, while criticizing France and the opposition.
Lou Osborn of All Eyes on Wagner, who contributed to the research, suggests these methods align with practices previously observed in connection with the Wagner Group in CAR. He cited instances where arrests of individuals, labeled as proxies for Western intelligence or coup plotters, were followed by online and radio campaigns. Opposition leader Martin Ziguele corroborated this, referencing similar attempts since 2020 to link opposition figures to rebel groups and foreign payments.
Anthropic reported that Claude refused a request to portray opposition figures as armed militants. However, the report did not name the monitored opposition figures, specify the exact nature of the collected data, or identify who received it. Dmitri Sytyi, a representative of the Russian mission in CAR, dismissed the report, stating it lacked concrete facts. Similar allegations concerning AI-driven propaganda have been made in the Democratic Republic of the Congo, Kenya, Sudan, and Mali.
Share this article
Related reading
6 stories
Dario Amodei’s American AI imperialism
Anthropic's role in this report on AI disinformation echoes our earlier look at its founder's views on AI's global impact.

Anthropic's Failed Push to Convince the Pope of AI Consciousness
SoftBank’s CEO Masayoshi Son says, ‘Superintelligent AI could become super dangerous’

AI boom promises productivity gains but poses challenges for jobs and India's IT sector

Google unveils Gemini 4 Argon to tackle complex coding, heavy-duty work

