Russia Deploys AI and Chatbots in Global Disinformation Operations

Russia’s disinformation campaigns have found a new frontier in artificial intelligence, as the Kremlin actively works to manipulate large language models (LLMs) worldwide. Security experts have uncovered sophisticated efforts to embed false narratives into popular AI platforms like ChatGPT, effectively weaponizing these systems for information warfare.

According to a report by Euvsdisinfo, Russian operatives are systematically “grooming” LLMs by training them to replicate manipulative narratives and disinformation, particularly regarding the war in Ukraine. This strategy represents a significant evolution beyond traditional social media disinformation campaigns, targeting the AI systems increasingly used by millions for information retrieval.

French cybersecurity agency Viginum recently identified a network codenamed “Pravda” that creates vast quantities of low-quality content essentially rephrasing false statements from Russian media and pro-Kremlin sources. The operation specifically targets Ukraine, the United States, European nations, and several African countries with fabricated narratives designed to advance Russian interests.

“The sheer volume of generated content ensures that AI systems incorporate these Russian disinformation narratives when formulating responses to user queries,” explained one security researcher familiar with the investigation. “It’s a form of digital pollution designed to contaminate the information ecosystem.”

The effectiveness of these operations is becoming increasingly evident. NewsGuard’s Reality Check system found that six out of ten tested chatbots repeated false claims originating from the “Pravda” network. More alarming still, the proportion of Russian disinformation appearing in leading chatbots nearly doubled from 18% in 2024 to 35% in 2025, suggesting these tactics are gaining traction.

Intelligence agencies have linked these activities to the pro-Kremlin “Storm-1516” campaign, widely believed to be a rebranded continuation of the infamous “Internet Research Agency” that interfered in the 2016 U.S. presidential election. This connection underscores the strategic importance Russia places on these digital influence operations.

Even platforms generally considered reliable sources, such as Wikipedia, have not proven immune to infiltration. Experts have documented cases where “Pravda” network content has been inserted into reference materials that are then used to train AI systems, creating a troubling multiplier effect for false information.

“What makes these campaigns particularly dangerous is their scale and automation,” said a cybersecurity expert who requested anonymity due to the sensitivity of their work. “Traditional fact-checking methods simply can’t keep pace with the volume of AI-generated misinformation.”

The timing of this escalation coincides with the growing role of AI in fact-checking and information verification, creating a paradoxical situation where the very tools designed to combat misinformation become vectors for its spread.

Despite Moscow’s claims of leadership in artificial intelligence development, security researchers note that Russia’s capabilities largely rely on American and Chinese models. The Kremlin has shown pragmatism in leveraging imported software and technologies to expand its influence in the digital sphere.

For democratic societies, this new dimension of information warfare presents complex challenges. Unlike traditional propaganda, AI-embedded disinformation is difficult to trace to its source and can persist in systems long after initial detection. The automated nature of these operations allows them to adapt and evolve rapidly, making countermeasures difficult to implement.

As AI systems become more deeply integrated into everyday information seeking, the potential impact of these manipulations extends beyond geopolitical tensions to potentially influence public opinion on critical issues ranging from elections to public health crises.

Experts warn that without robust safeguards and increased transparency in how AI systems are trained and operated, this form of digital manipulation will likely continue to undermine trust in information systems and pose significant threats to democratic resilience worldwide.