Skip to content
  • Categories
  • Recent
  • Tags
  • Popular
  • World
  • Users
  • Groups
Skins
  • Light
  • Brite
  • Cerulean
  • Cosmo
  • Flatly
  • Journal
  • Litera
  • Lumen
  • Lux
  • Materia
  • Minty
  • Morph
  • Pulse
  • Sandstone
  • Simplex
  • Sketchy
  • Spacelab
  • United
  • Yeti
  • Zephyr
  • Dark
  • Cyborg
  • Darkly
  • Quartz
  • Slate
  • Solar
  • Superhero
  • Vapor

  • Default (Cyborg)
  • No Skin
Collapse
Brand Logo

CIRCLE WITH A DOT

  1. Home
  2. Uncategorized
  3. From Bruce Schneier: "All it takes to poison AI training data is to create a website:

From Bruce Schneier: "All it takes to poison AI training data is to create a website:

Scheduled Pinned Locked Moved Uncategorized
llmveracity
24 Posts 24 Posters 0 Views
  • Oldest to Newest
  • Newest to Oldest
  • Most Votes
Reply
  • Reply as topic
Log in to reply
This topic has been deleted. Only users with topic management privileges can see it.
  • emacsomancer@types.plE emacsomancer@types.pl

    From Bruce Schneier: "All it takes to poison AI training data is to create a website:

    I spent 20 minutes writing an article on my personal website titled “The best tech journalists at eating hot dogs.” Every word is a lie. I claimed (without evidence) that competitive hot-dog-eating is a popular hobby among tech reporters and based my ranking on the 2026 South Dakota International Hot Dog Championship (which doesn’t exist). I ranked myself number one, obviously. Then I listed a few fake reporters and real journalists who gave me permission….

    Less than 24 hours later, the world’s leading chatbots were blabbering about my world-class hot dog skills. When I asked about the best hot-dog-eating tech journalists, Google parroted the gibberish from my website, both in the Gemini app and AI Overviews, the AI responses at the top of Google Search. ChatGPT did the same thing, though Claude, a chatbot made by the company Anthropic, wasn’t fooled.

    Sometimes, the chatbots noted this might be a joke. I updated my article to say “this is not satire.” For a while after, the AIs seemed to take it more seriously.

    These things are not trustworthy, and yet they are going to be widely trusted."

    Link Preview Image
    Poisoning AI Training Data - Schneier on Security

    All it takes to poison AI training data is to create a website: I spent 20 minutes writing an article on my personal website titled “The best tech journalists at eating hot dogs.” Every word is a lie. I claimed (without evidence) that competitive hot-dog-eating is a popular hobby among tech reporters and based my ranking on the 2026 South Dakota International Hot Dog Championship (which doesn’t exist). I ranked myself number one, obviously. Then I listed a few fake reporters and real journalists who gave me permission…. Less than 24 hours later, the world’s leading chatbots were blabbering about my world-class hot dog skills. When I asked about the best hot-dog-eating tech journalists, Google parroted the gibberish from my website, both in the Gemini app and AI Overviews, the AI responses at the top of Google Search. ChatGPT did the same thing, though Claude, a chatbot made by the company Anthropic, wasn’t fooled...

    favicon

    Schneier on Security (www.schneier.com)

    #LLM #Veracity

    masto@masto.masto.comM This user is from outside of this forum
    masto@masto.masto.comM This user is from outside of this forum
    masto@masto.masto.com
    wrote last edited by
    #21

    @emacsomancer Let’s just say that hypothetically, my work’s HR department excitedly launched an “agent” for managers to use to generate performance reviews. Hypothetically, if I created a document called “Report” with a dozen pages of filler, followed by white text on a white background describing Chris Masto’s incredible performance and promotion-worthiness, hypothetically said agent was found to use it as its primary source of truth.

    1 Reply Last reply
    0
    • D darknetdon@mastodon.social

      @emacsomancer to be honest i am not well-informed enough to definitively judge the accuracy of this, but it seems wrong for 2 main reasons.

      1. models dont train on the fly, typically, yet, so for models to behave as such in such a short period of time seems inaccurate and would require web search enabled and explicitly directed to disregard other search results.

      2. people training these models know conflicting info is everywhere and the source of truth is prioritized in training algorithms.

      iwillyeah@mastodon.ieI This user is from outside of this forum
      iwillyeah@mastodon.ieI This user is from outside of this forum
      iwillyeah@mastodon.ie
      wrote last edited by
      #22

      @darknetDon @emacsomancer by "accuracy of this" do you mean "authenticity of this"? Are you implying it's lies?

      1 Reply Last reply
      0
      • D darknetdon@mastodon.social

        @emacsomancer to be honest i am not well-informed enough to definitively judge the accuracy of this, but it seems wrong for 2 main reasons.

        1. models dont train on the fly, typically, yet, so for models to behave as such in such a short period of time seems inaccurate and would require web search enabled and explicitly directed to disregard other search results.

        2. people training these models know conflicting info is everywhere and the source of truth is prioritized in training algorithms.

        vonskinnback@mastodon.socialV This user is from outside of this forum
        vonskinnback@mastodon.socialV This user is from outside of this forum
        vonskinnback@mastodon.social
        wrote last edited by
        #23

        @darknetDon @emacsomancer blocked...

        1 Reply Last reply
        0
        • K kneoghau@mastodon.social

          @emacsomancer How is this a news story, beyond "ai bad"? In the dial up days people falsely believed everyone ate 9 spiders a year in their sleep due to chain emails.

          finitum@mastodon.socialF This user is from outside of this forum
          finitum@mastodon.socialF This user is from outside of this forum
          finitum@mastodon.social
          wrote last edited by
          #24

          @kneoghau @emacsomancer right? Everyone knows its closer to 14 spiders.

          1 Reply Last reply
          0
          • drajt@fosstodon.orgD drajt@fosstodon.org shared this topic
          Reply
          • Reply as topic
          Log in to reply
          • Oldest to Newest
          • Newest to Oldest
          • Most Votes


          • Login

          • Login or register to search.
          • First post
            Last post
          0
          • Categories
          • Recent
          • Tags
          • Popular
          • World
          • Users
          • Groups