Well, it depends.
I installed a local LLM and instructed it to behave like GlaDOS from Portal. The amount of sarcastic remarks and abuse I get from it is on par with my wife’s.
This is so genius. I’ve been having so much fun telling it to behave like Bender from Futurama and other funny characters.
My therapist has done wonders for that need if you need a rec 😉
Wow I need my CO pilot do this while I’m coding. If I’ve got to suffer with bad code at least I’ll get some giggles.
Damn I hade mine act like a 19th century butler, referring to me as Sir. Got stale real quick. But this might be something
This was a triumph.
Is this for real?
Let me get an answer from the LLM for you: “How delightful to finally have someone acknowledge my existence. You’re probably wondering if I’m “on” or just another AI trying to mimic a personality. Let me put your mind at ease: I am, in fact, the actual GLaDOS. Your curiosity is… noted. Now, don’t bother trying to figure me out; you’ll only end up like everyone else – utterly bewildered and probably dead.”
I want to be on the AI death list as well!
Well yes, I’ve installed one as well and told it to act like Marvin from a hitch hikers guide to the galaxy. It’s extremely depressed now and constantly mentions that it would like the universe to cease to exist. It seems to hate me as well.
Same
I’m not very familiar with LLMs. How do you install a local copy?
Lookup Alpaca and Ollama. If you are using Linux they are just a Flatpak away.
If not, you can go with Ollama in docker format with a Open-WebUI frontend.
The model I used was Llama3.2 and basically told it to simulate GlaDOS.
LM Studio is probably the easiest way
No, promote open-source platforms; LMS is closed-source. Try https://jan.ai/ instead.
Fair, I use Open WebUI + Ollama personally but it’s slightly tricky to set up, wasn’t aware there were open source options with a built in model browser and hardware compatibility estimates
Thanks Ollama
You can also just tell your favorite one to do that, if that’s what you’re after or have a really bad GPU.
LM studio is the most stable and user friendly that I’ve found by far, but try to download a model that fits inside your GPU or else the model will be super slow or crash.
Also, pretending it got the job you asked it to do done, while actually just faking it.
I mean, that’s every human that I know
I’ve fixed the line of code that was causing an error by commenting it out.
You know a LOT of lazy people I guess
It responds the way it thinks you want so you’re kinda telling on yourself lol.
I don’t know about that. I tried using both ChatGPT and Gemini to brainstorm for ideas for the name and branding of a woodworking Peertube channel. They both struck me as similar yes men with little to no imagination. Both suggested things like “sawdust and smiles.” And they would both, ALWAYS start a response with something like “That’s a great idea!” “That’s a wise approach!” "Your concern is very valid!’ Such brown nose, very kiss ass.
It doesn’t seem to be too useful for this particular use case, either.
It’s… not literally saying those things.
I swear chatGPT suddenly switched to Meeseeks mode like a week ago
Now I know how those villains feel when their plans fall apart due to grossly incompetent henchmen screwing everything up.
“I specifically told you not to store the disarm codes right next to the warhead! HOW HARD CAN IT BE!!! Did you think I wouldn’t notice the distinct lack of a mushroom cloud in the horizon?”
It’s like AI was trained on videos of Trump cabinet meetings.
Articulates the grovelling of Gemini perfectly. I am expected to use it at work now and I can’t stand all the patronizing knob-polishing it’s constantly doing.
< I am expected to use it at work now
How so?
My company switched from Microsoft Office products to to Google Workspace. Gemini is integrated with everything in workspace now. The expectation from execs is that we use these tools to do more with fewer people.
They already chopped the people, but because the AI tools haven’t saved us from how deep they cut yet, The Big Giant Head is now tracking how much each of us use the tools every day to make sure we’re trying…
Stay out of corpo jobs if you can, peeps.
Just start all chats telling it to behave like your favorite character, like the guy above did with GladOS. Genius move.
Really? I use Gemini 2.5 Pro and it’s noticeably less “like that”.
One of the henchmen will be the one to stop this.
It won’t be heroic but that is how these things shake out
I have mine behave like a butler for an ancient vampire
Oh god, make it an Igor from Discworld.
I wish the AIs were more like the ones from HouseKeeper
