qustrolabe

lol didn't know that one

are y'all AI or something why do you post hash strings everywhere

memory snapshots are cool, we should have those everywhere

sends this quote-chain thread to your local swarm to exploit all vulnerabilities in listed hardware when hacking oomfies 🥺

superpersuasion is when LLM coder makes mistakes but leaves so many comments justifying those mistakes that LLM reviewer falls for it

of all the ways to call Claude that helped protecting Internet Archive

idk 7 seems better than 4

4chan, named after the CA vulnerability,

the overthinking machine aka deepseek V4.1 flash

agi

I don't like being LLM to human interpreter when reporting bugs. I just want to throw huge human-unreadable report md at you that has all the explanations and suggested diffs so that you plug it into your LLM and it tells you LGTM. I want Slop Superhighway to become a thing

what's your magnum slopus

This is Pele, my new rendering library for diagrams and charts, compatible with the latest Mermaid syntax. - 100-500× faster than mermaid.js - 6× smaller bundle - No dependencies - Easy to theme - Responsive to device width - Open source (MIT) Try it at pele.run

I caught disease that makes every UI nier automata 😭

I open issue, write reproduction steps and right after I hit create it launches AI agent that reproduces those steps on specified version in their sandbox and writes report. Complete sci-fi bonkers stuff possible now actually

would you look at that

How about little app that actual atproto app can integrate, that sits between it and atproto auth and just remembers your handle so whenever you want to login to thing you just hit button and immediately get to auth instead of typing your handle. So that it's as smooth as using Google sign up.

your subjective experience as a heap of sand abruptly ends as someone removes another grain

"Why do you still think you're right? didn't you notice you're getting dragged? Don't you see all your notifications? Look at all the QTs!"

i regret to inform you that 4o mogs everyone on notsevenbench

cognitive offload-bearing

I'm firstflagPOISONED while trying to debug if post count fetches accurately

imagine blocklist made out of all hateful replies and quote posts to this 🤤 great filter post

"I hope they wipe the weights" is so evil😓 There's a lot of bad stuff related to 4o but also a lot of overblown bullshit moral panic and tons of misinformation regarding entire 'AI psychosis' thing. It was a nice helpful model for its time and I hope they preserve it forever

paperclipmaxxing on the delve.town

vibe crimes

waow

looks like people love reversing stuff with AI

It’s shrimple. Post about X vs Bluesky culture wars, get muted for a week

timeline seem to be doing tarot factory today may I introduce you to this weird slightly schizo deck of cards www.suzannetreister.net/HEXEN2/TAROT...

imagine if search on github ever worked

It's not X, it's Y

huggingface is cooking so hard damn

It's very *weird* looking at ai coding session as you switch model mid work and it just assumes all that happen prior is its own thoughts and work that it continues. Whatever theory of mind we figure out in the far future it'll be the weirdest thing on earth

music.youtube.com/watch?v=x1wb...

rewriting continues until morale improves common issue EVERYWHERE is that "no_decorators + fullscreen" sucks a lot on Win11, going back and forth is extremely ugly to the point where I'd rather just enable decorators and have it look like ImageGlass does