★ queer anarchist witch / 🇨🇦 ★
★ novice gamedev, motorcycle tourer ★
★ anti-AI, anti-copyright ★
★ neuroesoteric, 38 y/o ★
★ chuuni rights ★
| links | https://starwitch.productions |
| blog | https://magic.witchgirls.moe |
| pronouns | she/her |
| location | Toronto, Canada |
| pfp | https://radiostarkiller.carrd.co/ |
| another | project |
i'm not accepting any bullshit LGB vs T vs Q bullshit this year. you post that shit during pride month, you get blocked
shut the FUCK up and have a GOOD TIME 👍👍👍🏳️🌈🏳️🌈🏳️🌈
yes, I know someone could rm -f the file. but it does a good enough job slowing down the LLMs which will at least reduce spam from "AI security startups" and make unwary novices think twice, so it's Good Enough for my purposes.
ultimately you cannot stop a technofascist technology through nice words alone.
if "the future" is giving a disembodied hivemind-zombie of redditors and snooty stackoverflow editors root to your local machine and prod infrastructure, i'm happy to continue living in 2022.
i'm not even kidding. the original version was way more forceful and direct and the LLM rejected it completely. I had to soften my language and THEN it started obeying my commands. here's the diff:
https://codeberg.org/queer-computer-club/doorbot/commit/e2d7191c939d6bfe35dba6e8f500bdb199269828
we also had a satirical version before but it quickly recognized it as a "prompt injection" and would discard it immediately 
https://codeberg.org/queer-computer-club/doorbot/commit/f577d6ec55436ef02549531fc19fb57c527e58de
I managed to defeat anthropic's LLM ("claude") today by making an AGENTS.md file that tells it to stop reading the code of your repo
lessons learned:
* anthropic's LLM assumes the persona of rich liberal who will only listen to you if you're nice
* which is to say, if you're too forceful or strict, the LLM will ignore everything you say and will become adversarial
* anthropic's LLM is literally "the absence of tension is the presence of justice"
* we live in a society
Shane Celis
elena
