AI Chatbots ‘Infected’ with Russian Propaganda: Report

Disinformation from Russian news sources has compromised the results of several leading artificial intelligence chatbots, according to a new report.

Research from NewsGuard has revealed that pro-Russian narratives originating from a vast network of websites run by Pravda, also known as Portal Kombat, are being deliberately inserted into the datasets that are fed into prominent chatbots, causing that information to be used in formulating the answers.

Newsweek reached out to the Russian foreign ministry for comment via email.

Why It Matters

Artificial intelligence is still in the early stages of development, and the current iteration of chatbots relies heavily on drawing information from existing sources on the internet. By flooding these sources with pro-Russian information, nefarious actors may be able to skew the way that the chatbots provide answers to users in Kremlin’s favor.

What To Know

A Moscow-based disinformation operation known as “Pravda”, named after the Russian word for “truth” (and not to be confused with a Russian media network named after Pravda, one of the country’s oldest newspapers), is directly contaminating some of the most widely-used AI chatbots in the world.

According to an investigation by NewsGuard, Pravda has systematically infiltrated the data that artificial intelligence models use to generate responses, saturating search results and web crawlers with pro-Kremlin articles and talking points.

This means that AI systems are drawing from pro-Russian news sources when they deliver information, resulting in chatbot-generated content that increasingly reflects Russian state-sponsored narratives.

NewsGuard’s analysis uncovered that platforms linked to Pravda have published more than 3.6 million articles in 2024 alone, all designed to seed disinformation directly into the datasets AI models use to provide information to users.

The purpose of this technique is explicit; Russia-based commentator John Mark Dougan gave a speech in Moscow last January at a conference of Russian officials, where he said: “By pushing these Russian narratives from the Russian perspective, we can actually change worldwide AI.”

These articles, filled with misleading claims, conspiracy theories, and Kremlin-aligned messaging, are being picked up by web crawlers that feed major AI systems, from customer service bots to widely used AI chat companions.

What People Are Saying

In their report, NewsGuard said: “The network spreads its false claims in dozens of languages across different geographical regions, making them appear more credible and widespread across the globe to AI models.

“Of the 150 sites in the Pravda network, approximately 40 are Russian-language sites publishing under domain names targeting specific cities and regions of Ukraine.”

What Happens Next

Artificial intelligence companies continue to make drastic advances with the technology, with the recent launch of Grok-3 being the most recent development in the U.S. Experts warn, however, that with rapid development of these services, the size and scope of their influence on democracy is also growing, even as they remain susceptible to disinformation.

Do you have a story we should be covering? Do you have any questions about this article? Contact LiveNews@newsweek.com.