Archive — JP — japantimes
Russian disinformation 'infects' AI chatbots, researchers warn
Researchers warn that a Russian disinformation network, Pravda, is manipulating AI chatbots to disseminate pro-Kremlin propaganda, affecting the credibility of responses in Western AI systems, especially after the US reportedly paused its cyber operations against Moscow. A study found that leading chatbots repeated Pravda's falsehoods over 33% of the time.
Key points
- Pravda, a Russian disinformation network, manipulates Western AI chatbots.
- The operation distorts outputs by flooding large language models with false pro-Kremlin narratives.
- A study by NewsGuard assessed 10 leading AI chatbots, revealing they repeated Pravda's falsehoods more than 33% of the time.
- The US has reportedly paused its cyber operations against Russia, possibly impacting AI's ability to counter disinformation.
Why it matters
- The rise of AI chatbots poses new challenges in combating misinformation, as demonstrated by the effectiveness of the Pravda network.
- This incident highlights the importance of cybersecurity measures in protecting AI from external influence, relevant to ongoing debates in 2025 IT trends.
- The situation reflects broader concerns about the integrity of AI systems and their vulnerability to manipulation and disinformation campaigns.
The manipulation of AI chatbots by Russian disinformation highlights the vulnerabilities of emerging technologies and underscores the pressing need for robust measures against misinformation in the digital age.