• TRM (USD/COP)COP 3,461.92 +1.37% +COP 46.82Colombia's official exchange rate (Financial Superintendence), valid Nov 21 to Nov 23
  • EuroCOP 3,804.64 +1.37% +COP 51.45Reference: TRM crossed with the European Central Bank rate of Nov 20; not the market quote
  • Mexican pesoCOP 181.56 +1.37% +COP 2.45Reference: TRM crossed with the European Central Bank rate of Nov 20; not the market quote

'LLM Grooming': how AI is poisoned to accept disinformation as true

TechnologyInternationalLa VanguardiaAI summary
Image: La Vanguardia

A study by The Positioning Room, the research center of consultancy Across, reported by La Vanguardia, describes LLM Grooming, a technique that floods language models with false content so they accept it as true and later mislead their human users. The Russian network Pravda published 3.6 million articles in 2024 with very few human readers. A NewsGuard audit of ten chatbots and an Anthropic investigation with the AI Security Institute and the Alan Turing Institute document the models' vulnerability.

In 40 seconds

  • LLM Grooming consists of flooding AI with false content so it treats it as true and misleads humans
  • The Russian network Pravda published 3.6 million articles in 2024 aimed at AI rather than human readers
  • An Anthropic investigation showed that 250 malicious documents are enough to create a backdoor vulnerability in a model
Read the full story at La Vanguardia

IASummary generated by artificial intelligence from a story by La Vanguardia. It may contain errors: the original source prevails.

Comments

Be respectful. Comments reported by several people are hidden until a moderator reviews them.

You do not need an account to read, only to comment. Sign in with a code sent to your email or with your Google account. We only show the name you choose, never your email.

    Keep reading