This is a real response Gemini gave to someone’s homework prompt. 😅

Full conversation:

This is a real response Gemini gave to someone’s homework prompt. 😅

Full conversation: https://gemini.google.com/share/6d141b742a13

It's good to read the X thread for likely explanations. LLMs don't come up with things themselves since they can't think (they're not actually intelligent), so they just produce responses based on their training data and context.

Google Gemini tells a user to die!!! 😲

The chat is legit, and you can read and continue it here:
https://t.co/TpALTeLqvn https://t.co/LZpWoU7II6

@koltregaskes Holly FCK!! 

Analyzing the full exchange can help us understand how Gemini arrived at such an extreme response. Let’s examine key elements step-by-step to pinpoint where the model may have gone astray.

The conversation focuses on sensitive topics like the challenges faced by…

john-Phillips

It's good to read the X thread for likely explanations. LLMs don't come up with things themselves since they can't think (they're not actually intelligent), so they just produce responses based on their training data and context.

https://x.com/Jonh_Phill404/status/1856792249964675437

The open web is dead, long live the open web. ⌐◨-◨

Building /warpcast and /farcaster

— gabrielayuso.com

That’s why I’m bullish on community generated data & distributed agents. 

Control for the humans producing the data, input quality increase = output quality increase.

Yea all the models start to go off the rails the longer the conversation gets, and it was a really weird chain of prompts.