NPR and NewsGuard tested how AI chatbots handle foreign propaganda by posing 30 questions built from false narratives that originated in China, Iran, and Russia. The study found that popular chatbots such as ChatGPT and Gemini correctly pushed back on the false narratives roughly three-quarters of the time, outperforming traditional search engines. AI-generated summaries appearing at the top of search results performed worse than standalone chatbots, with effectiveness varying across Google, Bing, and DuckDuckGo. The researchers still recommend that users verify primary sources regardless of which tool they start with.
