How Easily Can Russian Propaganda Fool AI Models? A New Benchmark Finds Out
This story is only covered by news sources that have yet to be evaluated by the independent media monitoring agencies we use to assess the quality and reliability of news outlets on our platform. Learn more here.4 Articles
4 Articles
Researchers at the Institute of Estonian Language accused the French company Mistral AI that its models tended to reproduce "Russian propaganda narrations." All four versions of the "Mistral" study received low ratings in their ability to recognize and dissipate information that was considered "Kremlin theories" at the Estonian Institute. In the course of its work, the Institute tested 60 models of popular YI-chat-bots, including OpenAI and Anth…
A new study has exposed the weaknesses of Europe's leading AI model, Mistral, in Russian propaganda, showing how it often adopts Kremlin narratives without correction.
A new study has shown that Europe's most prominent artificial intelligence (AI) model is vulnerable to Russian propaganda. Researchers from the Institute of the Estonian Language asked 60 popular AI chatbot models a series of 75 questions in three languages. The aim was to see how they would spread Russian propaganda or be able to filter it out. They included four versions of Mistral AI, ChatGPT from OpenAI and Claude from Anthropic, writes the …
How easily can Russian propaganda fool AI models? A new benchmark finds out
The Institute of the Estonian Language has released a benchmark measuring how susceptible AI language models are to Russian propaganda. The article How easily can Russian propaganda fool AI models? A new benchmark finds out appeared first on The Decoder.
Coverage Details
Bias Distribution
- There is no tracked Bias information for the sources covering this story.
Factuality
To view factuality data please Upgrade to Premium
