So when we think about the Arctic, a common narrative from Russia is that it alone can develop the region, which will end US hegemony and warmongering.
Honestly? Maybe not wrong. But where we see malign intent is, e.g., laundering this through the #Pravda network responsible for #LLMGrooming.
So, the process we all knew was going to happen, the seeding of #ai models with #propaganda, now has a name: #llmgrooming
«Most large language models (LLMs) provide sources along with their analysis. But these models do not choose what sources to cite based on credibility. Rather, they choose based on availability.»
#LLMs #genAI #factChecking #propaganda #LLMGrooming
LLM grooming: Hundreds of English-language websites link to pro-Kremlin propaganda
[The Guardian] Thinktank says internet flooded with disinformation by Russia-aligned Pravda network, which many websites treat as credible
#llm #llmgrooming #grooming #propaganda #russia #pravda
#LLMGrooming nennen es McKenzie Sadeghi und Isis Blachez:
Keine direkte Manipulation, sondern die Beeinflussung unseres Weltbildes, indem die Informationsgrundlage
verändert wird, auf der wir entscheiden.
Mehr: www.newsguardrealitycheck.com/p/a-well-fun...
"Pohjoismaisten faktantarkistajien selvitys osoittaa, että tekoälybotit ChatGPT, Copilot ja Gemini käyttävät lähteinään venäläisiä Pravda-sivustoja, kun niille esitetään pohjoismaisilla kielillä kysymyksiä Venäjän propagandasta."
faktabaari.fi/fakta/venaja... #LLMgrooming #GenAI
#LLMGrooming as a New Threat: How Pro-Kremlin Networks Prepare Information for #AI / #pravdanetwork #disinformation #fakenews factcheck.by/eng/news/llm...
Astroturfing, but for AI.
«[…] mass quantities of well-aimed chaff can skew the answers on specific topics. For Russia, that is the war in Ukraine. But for a politician, it could be an opponent; for a commercial firm, it could be a competitor.»
#AI #LLM #LLMGrooming #astroturfing #propaganda #FIMI
Misinformation isn’t just a human problem anymore - it's being baked into AI models.
Russia’s Pravda network used LLM Grooming to flood AI chatbots, causing them to repeat pro-Kremlin disinformation.
guardos.ai/blog/2025/gr...
#AISecurity #Cybersecurity #GenAI #LLMGrooming #AIExploit #news
"Prawda manipuliert die Chatbots den Forschern zufolge, indem das Netzwerk die großen Sprachmodelle (Large Language Model, LLM) der KI-Anwendungen, aus denen die Antworten generiert werden, mit seinen Fakenews flutet."
#LLMGrooming
What do we mean by #LLMGrooming?
We at @americansunlight.org wrote about the #Pravda network – a Russian propaganda aggregator churning out 3.6M pieces of disinformation a year – and theorized that, because its websites are so unusable by humans, they were probably being used to poison LLMs.
1/
Three-step-thread on who, how and why they found out about this particularity of #LLMgrooming
#LLMGrooming is yet another tool that the Russian disinfo machine has added to its repertoire. This is NOT a political or partisan issue. Combined with the fact that fake AI-generated personas are more active than ever (see our report about the #EcoBoost network)... without action, we're in trouble.
🚨BIG NEWS🚨 @newsguard.bsky.social has found 33% of answers generated by leading GenAI models, in response to false narratives about Ukraine, mirror content recycled by the Pravda Network.
www.newsguardrealitycheck.com/p/a-well-fun...
We reported on the likelihood of #LLMgrooming just a week ago!
