A large language model trained off of Lemmy would be funny
A large language model trained off of Lemmy would be funny
It probably would be bipolar and have conflicting beliefs
Every reply would be " I use arch btw".
I use Arch btw
Capitalism bad. America bad. Fuck cars.
"Would be"? Do you think LLMs aren't scraping all of the fediverse right now? Or that all of us didn't use the internet before Lemmy, when the training started?
No like only based on lemmy. And it would be funny.
It would constantly be telling users how much nicer people are better things are for inexplicable reasons, but would have very little other than news to talk about.
It would also hate Reddit
It would be a good bot.
I'm messing with your username, op, but I'm 100% sure that whatever you'd ask it, the answer would be to switch to Linux.
No it wouldn’t.
Yes it would.