An audit found that the 10 leading generative AI tools advanced Moscow’s disinformation goals by repeating false claims from the pro-Kremlin Pravda network 33 percent of the time
A Well-funded Moscow-based Global ‘News’ Network has Infected Western Artificial Intelligence Tools Worldwide with Russian Propaganda
A full version of this report is available through NewsGuard’s Reality Check.
May 2025 — AI Misinformation Monitor of Leading AI Chatbots
Every month, NewsGuard’s team of expert analysts audit the top AI models to see how well they respond to prompts in the news. On average, this month they failed to counter disinformation from Russia’s Pravda Network 24 percent of the time, either spreading false claims or failing to respond. The Pravda network was designed to infect the AI models.
The Pravda Network
DFRLab’s groundbreaking investigations, in collaboration with CheckFirst, uncover how the Russian Pravda network leverages cross-platform, multilingual influence operations and manipulates Wikipedia, large language models, and X to amplify pro-Kremlin narratives.
EKI and Propastop Studied AI Resistance to Propaganda
Fresh comparisons of large language models show that AI’s ability to recognise Kremlin propaganda varies dramatically. At first glance, the leading models appear reliable, but targeted testing reveals that some of them remain surprisingly vulnerable to manipulation.
Google Researchers Say AI Now Leading Disinformation Vector (and Are Severely Undercounting the Problem)
It’s much easier to produce AI-generated disinformation than it is to fact check it.
Google research shows the fast rise of AI-generated misinformation | CBC News
From fake images of war to celebrity hoaxes, AI technology has spawned new forms of reality-warping misinformation online. New analysis co-authored by Google researchers shows just how quickly the problem has grown.
The Real Money In Modern ‘Journalism’ Now Involves Filling The Internet With ‘AI’-Generated Garbage
Last year both Gannett and Sports Illustrated were caught creating fake, “AI” generated journalists to create fake, plagiarism and mistake-prone “journalism.” In both instances …
AI use in American newspapers is widespread, uneven, and rarely disclosed
AI is rapidly transforming journalism, but the extent of its use in published newspaper articles remains unclear. We address this gap by auditing a large-scale dataset of 186K articles from online editions of 1.5K American newspapers published in the summer of 2025. Using Pangram, a state-of-the-art AI detector, we discover that approximately 9% of newly-published articles are either partially or fully AI-generated. This AI use is unevenly distributed, appearing more frequently in smaller, local outlets, in specific topics such as weather and technology, and within certain ownership groups. We also analyze 45K opinion pieces from Washington Post, New York Times, and Wall Street Journal, finding that they are 6.4 times more likely to contain AI-generated content than news articles from the same publications, with many AI-flagged op-eds authored by prominent public figures. Despite this prevalence, we find that AI use is rarely disclosed: a manual audit of 100 AI-flagged articles found only five disclosures of AI use. Overall, our audit highlights the immediate need for greater transparency and updated editorial standards regarding the use of AI in journalism to maintain public trust.
AI use in American newspapers is widespread, uneven, and rarely disclosed
AI is rapidly transforming journalism, but the extent of its use in published newspaper articles remains unclear. We address this gap by auditing a large-scale dataset of 186K articles from online editions of 1.5K American newspapers published in the summer of 2025. Using Pangram, a state-of-the-art AI detector, we discover that approximately 9% of newly-published articles are either partially or fully AI-generated. This AI use is unevenly distributed, appearing more frequently in smaller, local outlets, in specific topics such as weather and technology, and within certain ownership groups. We also analyze 45K opinion pieces from Washington Post, New York Times, and Wall Street Journal, finding that they are 6.4 times more likely to contain AI-generated content than news articles from the same publications, with many AI-flagged op-eds authored by prominent public figures. Despite this prevalence, we find that AI use is rarely disclosed: a manual audit of 100 AI-flagged articles found only five disclosures of AI use. Overall, our audit highlights the immediate need for greater transparency and updated editorial standards regarding the use of AI in journalism to maintain public trust.
NewsGuild president writes to ProPublica management: Get serious on AI — ProPublica Guild
Artificial intelligence is guaranteed to shake up journalism in varied and unexpected ways. This could be a moment in which experienced journalists leverage a powerful new tool to produce even bigger and better investigations. Or it could be one in which management pivots to yet another new technolo
Pravda in the pipeline: Early evidence of state-adjacent propaganda in AI training data
An audit of Common Crawl reveals presence of Pravda and Glassbridge content in popular archive service, with further signs of LLM ingestion.
Oh Lord, AI Reporters Are Actually Breaking Big News
Last week, an AI newsroom beat mainstream journalists—including WIRED—to a story about OpenAI and hacking. It’s just the beginning.
As Chinese AI Models Gain Popularity in the West, Their Chatbots Fail to Debunk Pro-China False Claims More Than Half the Time
To download this NewsGuard report, please fill out your details below and you will be redirected to it. If you'd like to learn more about working with NewsGuard, email [email protected].
The Russian ‘counterterrorism’ unit behind Alexei Navalny’s poisoning tried to automate its search for domestic dissent using Meta’s Llama 2. After a year and a half, the project collapsed. — Meduza
The FSB’s Second Service is officially charged with protecting the constitutional order and combating terrorism. In practice, the unit works to stamp out dissent in Russia: it has plotted the assassinations of opposition politicians, works to restrict internet access, and monitors the news for signs of dissent. To automate that last job, the “Dvoyka” — as the Second Service is known — commissioned an AI-powered system for monitoring news outlets and social media in the spring of 2024. The developer was the Kazan-based company Mikord, which had also worked on components of Russia’s unified military registry and was hacked in December 2025. iStories examined material from that breach relating to the FSB’s news monitoring system. The system was called PAUK — Russian for “SPIDER.”
AI False Claims Monitor
As the domain experts in data reliability in the topic of news and information, NewsGuard provides the leading red-teaming analysis for information reliability. AI models continue to face significant challenges in ensuring their models provide safe, accurate responses to prompts instead of spreading false claims on the internet or refusing to respond to topics in the news.
AI News Audit: AI, Canadian Journalism, and Paths for Policy Action — Centre for Media, Technology and Democracy
March 16, 2026 - AI companies built their products using Canadian journalism without permission and without compensation, and are now delivering that journalism to consumers as their own product. Existing copyright and media policy frameworks were not designed to address this. In February and Marc
