dkppunk comments on Everything's fine. Nothing to see here.

submitted 6 months ago* by mayabuttreeks@lemmy.ca to c/fuck_ai@lemmy.world

link to archived Reddit thread; original post removed/deleted

you are viewing a single comment's thread
view the rest of the comments

[–] dkppunk@piefed.social 13 points 6 months ago (1 child)

AIs aren't trained to produce correct answers, but seemingly correct answers

I prefer to say “algorithmically common” instead of “seemingly correct” but otherwise agree with you.

parent

[–] apftwb@lemmy.world 1 point 6 months ago* (1 child)

I use "mathmatical approximations of correct answers"

parent

[–] merc@sh.itjust.works 3 points 6 months ago

But that's wrong. It's not trained on correct answers. It's trained on whatever happens to be out there in the world.

It's mathematical approximations of words that are likely to be found near that question.

parent

lemmy:	0.19.20
mlmym (mschae):	26.3.4