Anthropic's new report reveals that Russia used the AI model Claude to support disinformation campaigns in the Central African Republic. The campaigns targeted opposition politicians and promoted pro-Russian narratives.
TL;DR
- Anthropic's report details Russia's use of Claude AI for propaganda in the Central African Republic and other African countries.
- The campaigns involved monitoring opposition figures and creating pro-Russian content, according to Anthropic.
- Anthropic shut down these operations by deleting involved accounts and enhancing detection systems.
What happened
Anthropic's report, titled 'Detecting and Countering Misuse of AI,' published in September, outlines how Russia used the AI model Claude to support disinformation campaigns in the Central African Republic. The report claims that Claude was primarily used to produce content for Radio Lengo Songo, a radio station funded by Russia, according to the investigative collective All Eyes on Wagner.
The content created by Claude was pro-Russian, portraying the Central African government and the Africa Corps mercenaries positively, while criticizing France and the opposition. The report also mentioned that Claude was prompted to portray opposition figures as armed militants, but the AI model refused the request.
Anthropic identified similar AI-driven influence operations in the Democratic Republic of the Congo (DRC), Kenya, Sudan, and Mali. In the DRC, 318 articles were published by a network of about 70 fake news sites and distributed by more than 250 fictitious accounts on X. In Kenya, Claude was allegedly used to generate batches of 50 messages supporting the government and opposing the opposition.
In Sudan, AI was used to create fabricated witness statements for the United Nations, and in Mali, an adviser with ties to intelligence services used Claude to develop a platform for monitoring telecommunications. Anthropic discovered and shut down these operations between December 2025 and August 2026 by deleting the accounts involved and strengthening its detection systems.
Why it matters
This report highlights the growing use of AI models like Claude in geopolitical influence operations, raising concerns about the potential misuse of AI in spreading disinformation and propaganda. For AI/ML developers, this underscores the importance of building robust detection and prevention mechanisms within AI models to counter such misuse.
Startup founders and tech investors should be aware of the ethical implications and potential regulatory scrutiny that may arise from the misuse of AI in political campaigns. This could lead to increased demand for AI security and ethical AI solutions, presenting new opportunities and challenges for the industry.
The competitive angle here is that Anthropic's proactive approach in detecting and countering the misuse of Claude sets a precedent for other AI companies. It demonstrates the importance of transparency and accountability in AI development and deployment, which could become key differentiators in the market.
Key facts
- Anthropic's report, 'Detecting and Countering Misuse of AI,' was published in September 2026.
- Claude AI was used to produce pro-Russian content for Radio Lengo Songo in the Central African Republic.
- Similar AI-driven influence operations were identified in the DRC, Kenya, Sudan, and Mali.
- In the DRC, 318 articles were published by a network of 70 fake news sites and distributed by 250 fictitious accounts on X.
- Anthropic shut down these operations by deleting involved accounts and enhancing detection systems between December 2025 and August 2026.
- Russia denies the allegations, stating that the report lacks concrete facts and evidence.
- Anthropic co-founder and CEO Dario Amodei emphasized the importance of transparency and accountability in AI development.
Context
The use of AI in political influence operations is not new, but the scale and sophistication highlighted in Anthropic's report are alarming. This report comes at a time when AI technologies are becoming more accessible and powerful, making it easier for malicious actors to leverage them for disinformation campaigns.
The broader AI landscape is increasingly focused on ethical considerations and the responsible use of AI. Companies like Anthropic are at the forefront of developing frameworks and technologies to detect and prevent the misuse of AI, setting a standard for the industry.
For developers and startups, this report serves as a reminder of the importance of building ethical considerations into AI development from the ground up. It also highlights the need for continuous monitoring and improvement of AI systems to adapt to emerging threats and misuse scenarios.
