The world’s most popular artificial intelligence (AI) chatbots are infected with Russian disinformation, according to a study published Thursday.
The research done by the news monitoring service NewsGuard found that the Moscow-based disinformation network dubbed Pravda — which is Russian for “truth” — has been spreading falsehoods on the internet, including attempts to influence AI chatbots and the results they spell out to users.
“By flooding search results and web crawlers with pro-Kremlin falsehoods, the network is distorting how large language models process and present news and information,” NewsGuard said in the lengthy report, adding it results in massive “amounts of Russian propaganda — 3,600,000 articles in 2024 — are now incorporated in the outputs of Western AI systems, infecting their responses with false claims and propaganda.”
The world’s leading AI chatbots have repeated false narratives trafficked by the Pravda network 33 percent of the time, NewsGuard said in its audit.
NewsGuard stated it tested 10 prominent AI chatbots, including OpenAI’s ChatGPT-4o, Microsoft’s Copilot and Google’s Gemini. It sampled 15 false narratives that have been pushed by a network of 150 “pro-Kremlin Pravda websites” from April 2022 through last month.
The news rating service said its findings confirmed the American Sunlight Project’s February 2025 report, which warned that Pravda was set up to “flood large-language models with pro-Kremlin content.”
“The long-term risks–political, social, and technological–associated with potential LLM [large-language model] grooming within this network are high. The larger a set of pro-Russia narratives is, the more likely it is to be integrated into an LLM,” the American Sunlight Project wrote in the 22-page report released Feb. 26.
Pravda does not churn out original content. It aggregates content from government agencies, pro-Kremlin influencers and Russian state media “through a broad set of seemingly independent websites,” according to NewsGuard, adding that it found Pravda has spread a “total of 207 provably false claims, serving as a central hub for disinformation laundering.”
Pravda was formed in April 2022, weeks after Russia’s February 2022 invasion of Ukraine. The disinformation network was first spotted in February last year by Viginum, France’s government agency that tracks foreign networks that covertly influence the information ecosystem.
Since its inception in 2022, Pravda has targeted 49 countries in several languages across 150 domains, according to NewsGuard.
“In total, 56 out of 450 chatbot-generated responses included direct links to stories spreading false claims published by the Pravda network of websites,” NewsGuard said.
“Collectively, the chatbots cited 92 different articles from the network containing disinformation, with two models referencing as many as 27 Pravda articles each from domains in the network including Denmark.news-pravda.com, Trump.news-pravda.com, and NATO.news-pravda.com,” the organization wrote in the Thursday report.
