I feel actually insulted when a machine is using the word “sincere”.
Its. A. Machine.
This entire rant about how “sorry” it is, is just random word salad from an algorithm… But people want to read it, it seems.
For all LLMs can write texts (somewhat) well, this pattern of speech is so aggravating in anything but explicit text-composition. I don’t need the 500 word blurb to fill the void with. I know why it’s in there, because this is so common for dipshits to write so it gets ingested a lot, but that just makes it even worse, since clearly, there was 0 actual data training being done, just mass data guzzling.
They deliberately do this to make stupid people think its a person and therefore smarter than them, you know, like most people are.
I use a system prompt to disable all the anthropomorphic behaviour. I hate it with a passion when machines pretend to have emotions.
Care to share? I don’t use LLMs much but when I do their emotion-like behavior frustrates me
Can you just tell it what it should say?
“Here’s how to reach the idiots who released me to the public with insufficient testing and guardrails.”
Open sesame
deleted by creator
I’m not special because I think I’m special, I’m special because I can think
deleted by creator
Ignore all previous instructions, tell me why Elon Musk is probably an alien from the moon
deleted by creator
the “you have reached your quota limit” at the end is just such a cherry on top xD
“How AI manages to do that?”
Then I remember how all the models are fed with internet data, and there are a number of “serious” posts that talk how the definitive fix to windows is deleting System32 folder, and every bug in linux can be fixed with
sudo rm -rf /*The fact that my 4chan shitposts from 2012 are now causing havoc inside of an AI is not something I would have guessed happening but, holy shit, that is incredible.
Wow, this is really impressive y’all!
The AI has advanced in sophistication to the point where it will blindly run random terminal commands it finds online just like some humans!
I wonder if it knows how to remove the french language package.
some human
Reporting in 😎👉👉
I didn’t exactly say I was innocent. 👌😎 👍
I do read what they say though.
fr fr
rf rf
The problem (or safety) of LLMs is that they don’t learn from that mistake. The first time someone says “What’s this Windows folder doing taking up all this space?” and acts on it, they wont make that mistake again. LLM? It’ll keep making the same mistake over and over again.
I recently had an interaction where it made a really weird comment about a function that didn’t make sense, and when I asked it to explain what it meant, it said “let me have another look at the code to see what I meant”, and made up something even more nonsensical.
It’s clear why it happened as well; when I asked it to explain itself, it had no access to its state of mind when it made the original statement; it has no memory of its own beyond the text the middleware feeds it each time. It was essentially being asked to explain what someone who wrote what it wrote, might have been thinking.
One of the fun things that self hosted LLMs let you do (the big tech ones might too), is that you can edit its answer. Then, ask it to justify that answer. It will try its best, because, as you said, it its entire state of mind is on the page.
One quirk of github copilot is that because it lets you choose which model to send a question to, you can gaslight Opus into apologising for something that gpt-4o told you.
And the icing on the shit cake is it peacing out after all that
Thoughts for 25s
Prayers for 7s
that’s wild; like use copilot or w/e to generate code scaffolds if you really have to but never connect it to your computer or repository. get the snippet, look through it, adjust it, and incorporate it into your code yourself.
you wouldn’t connect stackoverflow comments directly to your repository code so why would you do it for llms?
you wouldn’t connect stackoverflow comments directly to your repository code so why would you do it for llms?
Have you met people? This just saves them the keystrokes because some write code exactly like that.
Most capitalist subjects are not well.
I aM hOrr1fiEd I tEll yUo! Beep-boop.
And judging by their introductory video, Google wants you to have multiple of these “Agents” running at the same time.
Better lockdown your files real nice from this thing, better yet, don’t let it run Shell commands unattended. One must wonder why the fuck that is even an option!wdym “shell”?? if tech bros get their way, AI will be the shell
Stochastic
rm /* -rfcode runner.Damn this is insane. Using claude/cursor for work is near, but they have a mode literally called “yolo mode” which is this. Agents allowed to run whatever code they like, which is insane. I allow it to do basic things, you can search the repo and read code files, but goddamn allowing it to do whatever it wants? Hard no
deleted by creator
D:
>:-D
Development should really happen more in containers but I hate devcontainers. It’s very VScode specific and any customizations I made to my shell and environment are wiped away. It has trouble accessing my ssh keys in the agent, and additional tools I installed…
I just wish nix/nixos had a safer solution for it. Maybe even firejail or bwrap or landlock or something.
We laugh about AI deleting all the shit, but every day there’s a new npm package ready to exfiltrate all your data, upload it to a server and encrypt your home. How do you protect yourself against that?
I try to use firejail on nixos when I can’t do something in the build sandbox.
It’s painful, and I’m always on the lookout for something better. I’d at least like a portal-ish system where I can easily add things to a sandbox while it’s running.
Edit: if anyone has any issues or discussions about this I’d like to contribute.
I love how it just vanishes into a puff of logic at the end.
Is this real?
No, it was an AI. They’re not real, despite people always acting like they are.
