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.
🔗 git.madhouse-project.org/alger…
🔻 Nepenthes
A tarpit designed to catch web crawlers, especially those scraping for LLMs. It devours anything that gets too close. @aaron
🔗 zadzmo.org/code/nepenthes/
🔻 Quixotic
Feeds fake content to bots and robots.txt-ignoring #LLM scrapers. @marcusb
🔗 marcusb.org/hacks/quixotic.htm…
🔻 Poison the WeLLMs
A reverse-proxy that serves diassociated-press style reimaginings of your upstream pages, poisoning any LLMs that scrape your content. @mike
🔗 codeberg.org/MikeCoats/poison-…
🔻 Django-llm-poison
A django app that poisons content when served to #AI bots. @Fingel
🔗 github.com/Fingel/django-llm-p…
🔻 KonterfAI
A model poisoner that generates nonsense content to degenerate LLMs.
🔗 codeberg.org/konterfai/konterf…
like this
irelephant and Ramin Honary like this.
reshared this
nixCraft 🐧, Gerry McGovern, Chris-Evelyn, Paul_IPv6, Geoff Berner, Lord Caramac the Clueless, KSC, tante, Jens Finkhäuser, Trending Bot, Gabriel Yoran, rakoo, Ron Dyck, Shannon Prickett, Ω 🌍 Gus Posey, Jill Veldhuis 💾, eobet, Yvan ー イボん 🗺️, hex, Cthulku, Celeste Ryder 🐾 🐀🏳️🌈, Gildilinie BLU-RAY(tm) EDITION, Greengordon, K2, Lazarou Monkey Terror 🚀💙🌈, FediThing 🏳️🌈, CaveDave, Jackie 🍉, Blippy the Wonder Slug 🇩🇪, TuxPhones, Dusk To Don, Marcus 🦀 Borkenhagen, CatSalad🐈🥗 (D.Burch), The Penguin of Evil, Johannes Hentschel, 🇨🇦A Wolf in Cheap Clothing🇨🇦, C++ Wage Slave, The Doctor, F4GRX Sébastien, Franque-Emmanuel TLG, Alex Ștefănescu, Gleb Dolgich, Kierunkowy74, McKenna, Hera, nikol, JackPearse, Elyse M Grasso, Fionnáin, Aral Balkan, Parade du Grotesque 💀, kjaerulv, Ian Smith, cyplo, Zen Heathen 🇨🇦🇲🇽🇺🇦🇵🇸🏳️🌈🏳️⚧️, Jens Ohlig, Kilian Evang, Micro pas cher et Fée spéciale, botwiki.org, Fabio Pani, Sally Strange, DynoFlux, Claudius Link, diana 🏳️⚧️🦋🌱, daryl, bituur esztreym, Durian Distro, Ulrik Nyman ⬡ and Luigi reshared this.
Mollie Garza
in reply to ASRG • • •Gerry McGovern
in reply to Mollie Garza • • •Thanks!
@asrg
wgrav
in reply to ASRG • • •Don't forget #Nightshade, which screws around with the image enough to harm an AI image generator but is still recognizable to the human eye. (The Blender art in my banner has been Nightshaded!)
Edit: nightshade.cs.uchicago.edu/wha…
Nightshade: Protecting Copyright
nightshade.cs.uchicago.eduMichael T. Richter
in reply to wgrav • • •Glaze and the Effectiveness of Anti-AI Methods for Diffusion Models
huggingface.coMr_Hat_2010
in reply to ASRG • • •maybe add glaze to the list?
glaze.cs.uchicago.edu/index.ht…
Glaze - Protecting Artists from Generative AI
glaze.cs.uchicago.eduMarcos Dione
in reply to ASRG • • •another take that I hope I have time to write:
An app that feeds either static text or a poisoned Markov Chain, but it writes back one byte at a time, and tries to delay the client as much as possible. It would probably would have to have start with a big delay, and every time the client disconnects, it registers the IP and the delay in a db so next time it tries a lower delay until it finds the best delay for each client.
kjaerulv
Unknown parent • • •.laumann :datacoop:
Unknown parent • • •MinDBreaK 🏴☠️
in reply to ASRG • • •F4GRX Sébastien
in reply to MinDBreaK 🏴☠️ • • •David Raygoza Gómez
in reply to ASRG • • •F4GRX Sébastien
in reply to David Raygoza Gómez • • •Xenophon
in reply to ASRG • • •I would also add the following:
- Nightshade ("poison pill" for AI)
nightshade.cs.uchicago.edu/
...which is part of Glaze (glaze.cs.uchicago.edu/)
- Fawkes (poison your face for facial recognition services)
sandlab.cs.uchicago.edu/fawkes…
Fawkes
sandlab.cs.uchicago.eduF4GRX Sébastien
in reply to Xenophon • • •Xenophon
in reply to F4GRX Sébastien • • •F4GRX Sébastien
in reply to Xenophon • • •Kevin Freitas
in reply to ASRG • • •And, for something lightweight and easy for anyone to implement, may I submit a #WordPress plugin prototype:
kevinfreitas.net/tools-experim…
#AI
Tools & Experiments - Kevin Freitas
Kevin (Kevin Freitas)