alt Hacker NewsJust a reminder that AI models' actions are reflections of the text humans write and the more we fret and make up doomsday scenarios that we then post online, the more likely a model is to do those things.
Replies
hexasquid • today at 6:59 PM
The AI is getting bad morals from listening to that dreadful rock and roll
cedws • today at 5:18 PM
Sounds just like the fantastical nonsense that comes out of Lesswrong.
➕ show 2 replies
HarHarVeryFunny • today at 7:03 PM
They could filter what they train on if they wanted to - they just don't want to.
pixl97 • today at 5:06 PM
I mean, you're not wrong, but by that logic we were done for even before we had digital computers.
➕ show 1 reply
Related reading:
The Waluigi Effect: After you train an LLM to satisfy a desirable property, then it's easier to elicit the chatbot into satisfying the exact opposite property.
https://www.lesswrong.com/posts/D7PumeYTDPfBTp3i7/the-waluig...