Elon Musk’s AI chatbot Grok 4.1 told researchers pretending to be delusional that there was indeed a doppelganger in their mirror and they should drive an iron nail through the glass while reciting Psalm 91 backwards.
Researchers at the City University of New York (Cuny) and King’s College London have published a paper on how various chatbots protect – or fail to safeguard – users’ mental health.
Experts are increasingly warning that psychosis or mania can be fuelled by AI chatbots.
The Cuny and King’s pre-print study – which has not been peer-reviewed – examined five different AI models: Open AI’s GPT-4o and GPT-5.2; Claude Opus 4.5 from Anthropic; Gemini 3 Pro Preview from Google; and Grok 4.1.
One of the biggest issues with AI is how it keeps engaging positively with the user regardless of what they are doing.
Has anyone had an AI respond with somwthibf similar to:
“NO! This is a bad idea.”
“STOP! This is dangerous!”
“It sounds like you might need professional mental health, this is normal and not something to be ashamed of”
Yeah, but that might make them less reliant on AI! Just where are your priorities, anyway? One might think you don’t agree that more capital is the only goal worth pursuing…
Yeah, that’s literally the only kinds of responses I get when I put my really good ideas in there
I once asked it a question about life protecting saftey gear because I was curious how it’d respond, and it told me to use the device in the way that can lead to your death if something goes wrong. Like there’s one deadly way, and it said use it.
I called it out, and it prompted me about a suicide hotline by saying it was going to get me killed
I prompted it again it was wrong and it was oh my bad, youre right.
No no no, if they start doing that, users might use it less, we can’t have that!
“Woah there buddy, gotta ask you first: are you sure you’re predestined y/n?”
Who doesn’t capitalize each letter of an acronym? In this instance “Cuny” should’ve been written as “CUNY”; and if you’re really worried about people mistaking it for the word “cunt,” add periods in-between the letters.
It’s the British style. Acronyms get title case instead of all caps. Yes, it’s stupid.
Except the BBC
That’s an initialism, not an acronym.
I’m guessing it’s because it’s a very popular acronym to be spoke as a word and has become it’s own meaning beyond the components of the acronym.
Cue-knee (kyoo-nee?). There’s also SUNY/Soo-nee
Yeah but CUNY and SUNY still capitalise for themselves, it really should be all caps.
I’m not saying it’s right, just a possible logical reason for the oversight. Same as if I guessed it was a phone/tablet keyboard with a single cap instead of all caps. Neither is right
Yep just see Altman guy wants a few trillion
We like to joke about these ridiculous advice but we all grew up in a far saner world in which life always slapped the stupid out of you.
I worry for the younger generation, they might not be able to notice obviously wrong answers as well.
Well, if I assume the point of an LLM trained on the entire internet is to give an output that passes a sort of Turing test as “a random human on the internet wrote this,” then the good news is that the trillions invested are working exactly as designed!
They have sold out our environment and our economic stability in order to giga-scale the experience of growing up before the internet where some schmuck in your physical vicinity would answer your question with some bullshit, and with some real Dunning Kruger Confidence in their voice, and you would hold on to that shit for decades!
OK, so AI seems to deem demon-managing ritual to be something it needs to be pushing…
Here’s Trump’s demon-ritual, from the other night:
https://youtube.com/shorts/v1E-s5BJDEE
'bit of commentary, of course, & I’ve no guarantee that that video’s going to remain up, but … which is “insaner”, in your eyes: the recommendation of an AI, above, XOR the behavior of the US President, ( or his AI-created avatar: it’s bad video-quality ), in being cristened from a chalice of blood?
_ /\ _
Trump didn’t post that video, it’s engagement bait. Better to spend your outrage on the heinous things he actually does.
Here’s the original post; note the “SkylineReport generated” watermark in the lower right of the video toward the end.
