Russia deployed AI-powered disinformation campaigns including deepfakes, fake news websites, and coordinated bot networks to influence Germany's 2025 federal election, spreading false narratives about politicians and undermining democratic processes.
Russian intelligence services launched extensive AI-enhanced disinformation campaigns targeting Germany's February 23, 2025 federal election. The operation, codenamed 'Storm-1516' and linked to the 'Doppelganger' network, involved over 100 fake news websites managed by John Mark Dougan, a former U.S. police officer living in Russia. The campaign used AI tools like ChatGPT and DALL-E to generate fake articles, deepfake videos, and manipulated audio content. Specific disinformation included fabricated corruption scandals involving German politicians like Robert Habeck and Annalena Baerbock, false claims about migration agreements, and AI-generated videos showing fake pro-AfD demonstrations. The operation coordinated across multiple platforms including X, Telegram, Facebook, and TikTok, with bot networks amplifying content to millions of users. German intelligence agencies identified sharp increases in coordinated posts, with activity spiking to over 3,000 posts per day in late January 2025. The campaign targeted mainstream pro-Ukraine politicians while promoting the far-right Alternative for Germany (AfD) party, which was also endorsed by Elon Musk. German authorities confirmed the operation aimed to undermine trust in democratic institutions and influence voter preferences ahead of the election.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Using AI systems to conduct large-scale disinformation campaigns, malicious surveillance, or targeted and sophisticated automated censorship and propaganda, with the aim of manipulating political processes, public opinion, and behavior.
AI system
Due to a decision or action made by an AI system
Intentional
Due to an expected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed