ASRG (@asrg@tldr.nettime.org)
tldr.nettime.org
Attached: 1 image Sabot in the Age of AI Here is a curated list of strategies, offensive methods, and tactics for (algorithmic) sabotage, disruption, and deliberate poisoning. 🔻 iocaine The deadliest AI poison—iocaine generates garbage rather than slowing crawlers. 🔗 https://git.madhouse-project.org/algernon/iocaine 🔻 Nepenthes A tarpit designed to catch web crawlers, especially those scraping for LLMs. It devours anything that gets too close. @aaron@zadzmo.org 🔗 https://zadzmo.org/code/nepenthes/ 🔻 Quixotic Feeds fake content to bots and robots.txt-ignoring #LLM scrapers. @marcusb@mastodon.sdf.org 🔗 https://marcusb.org/hacks/quixotic.html 🔻 Poison the WeLLMs A reverse-proxy that serves diassociated-press style reimaginings of your upstream pages, poisoning any LLMs that scrape your content. @mike@mikecoats.social 🔗 https://codeberg.org/MikeCoats/poison-the-wellms 🔻 Django-llm-poison A django app that poisons content when served to #AI bots. @Fingel@indieweb.social 🔗 https://github.com/Fingel/django-llm-poison 🔻 KonterfAI A model poisoner that generates nonsense content to degenerate LLMs. 🔗 https://codeberg.org/konterfai/konterfai
  • arsCynic@beehaw.orgEnglish
    5·
    13 hours ago

    Stupidly trivial question probably, but I guess it isn’t possible to poison LLMs on static websites hosted on GitHub?

    • Luna@lemdro.idEnglish
      5·
      8 hours ago

      You can make a page filled with gibberish and have a display: none honeypot link to it inside your other pages. Not sure how effective would that be though

    • -dsr-@awful.systemsEnglish
      3·
      10 hours ago

      Sure, but then you have to generate all that crap and store it with them. Preumably Github will eventually decide that you are wasting their space and bandwidth and… no, never mind, they’re Microsoft now. Competence isn’t in their vocabulary.