HomeAI · summaries
Storia in 4 fonti
Mistral AI models score below 40% in detecting Russian propaganda, new benchmark reveals
Estonia's EKI benchmark tested 60 AI models on Russian propaganda resistance. All four Mistral models scored below 40%, ranking in the bottom third as the
Confronto fonti
4 prospettive sulla stessa storiaTimeline cronologica
- ·
cryptobriefing.com
Mistral's chatbot repeats Russian disinformation half the time, audit finds
NewsGuard audit finds Mistral AI's Le Chat repeats Russian and state-sponsored disinformation 50% of the time in English and 56.6% in French.
- ·
the-decoder.com
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.
