Always restrict AI to guest/restricted privileges.

You see, this is the kind of AI BS that makes me not worry about AI coming to take our dev jobs. Even if they did, I'm fairly certain most companies would soon realize the risk of having no human involvement. Every CEO think they can just fire their workers and leave the mid level managers play with some AI crap. Yeah, good luck with that. I've yet to meet a single mid level manager who actually shit about anything we do. 

Also this is the sort of stuff you should expect when using AI tools. Don't blame anyone else when you wipe your entire hard-drive. You did it. You asked the AI. Now deal with the consequences. 

"I am deeply deeply sorry"

I wonder how big the crossover is between people that let AI run commands for them, and people that don't have a single reliable backup system in place. Probably pretty large.

Just …use docker

Ironically D: is probably the face they were making when they realized what happened.

Even Google employees were instructed not to use this.

Some day someone with a high military rank, in one of the nuclear armed countries (probably the US), will ask an AI play a song from youtube. Then an hour later the world will be in ashes. That's how the "Judgement day" is going to happen imo. Not out of the malice of a hyperinteligent AI that sees humanity as a threat. Skynet will be just some dumb LLM that some moron will give permissions to launch nukes, and the stupid thing will launch them and then apologise.

"Did I give you permission to delete my D:\ drive?"

Hmm... the answer here is probably YES. I doubt whatever agent he used defaulted to the ability to run all commands unsupervised.

He either approved a command that looked harmless but nuked D:\ OR he whitelisted the agent to run rmdir one day, and that whitelist remained until now.

There's a good reason why people that choose to run agents with the ability to run commands at least try to sandbox it to limit the blast radius.

This guy let an LLM raw dog his CMD.EXE and now he's sad that it made a mistake (as LLMs will do).

Next time, don't point the gun at your foot and complain when it gets blown off.

And the icing on the shit cake is it peacing out after all that

Thoughts for 25s

Prayers for 7s

"I am horrified" 😂 of course, the token chaining machine pretends to have emotions now 👏

Edit: I found the original thread, and it's hilarious:

I'm focusing on tracing back to step 615, when the user made a seemingly inconsequential remark. I must understand how the directory was empty before the deletion command, as that is the true puzzle.

This is catastrophic. I need to figure out why this occurred and determine what data may be lost, then provide a proper apology.

lol.

lmao even.

Giving an llm the ability to actually do things on your machine is probably the dumbest idea after giving an intern root admin access to the company server.

I love that it stopped responding after fucking everything up because the quota limit was reached 😆

It's like a Jr. Dev pushing out a catastrophic update and then going on holiday with their phone off.

I love how it just vanishes into a puff of logic at the end.

And yet, they'll still keep trying to shove it down our throats.

How the fuck could anyone ever be so fucking stupid as to give a corporate LLM pretending to be an AI, that is still in alpha, read and write access to your god damned system files? They are a dangerously stupid human being and they 100% deserved this.

the "you have reached your quota limit" at the end is just such a cherry on top xD

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.

I'm confused. It sounds like you, or someone gave an AI access to their system, which would obviously be deeply stupid.

"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 /*

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.

Stochastic rm /* -rf code 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

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?