Jyunmi Hatcher

It just proves that the original poster is always right.

All right, well, before I go into this, I want to welcome Carl.

Carl, thanks for hopping on.

But to me, what this sounds like is AI is doing exactly what it was designed for way back in the day as a mirror to ourselves.

People like to be validated on their opinion.

People like to hear that, oh, you are right.

People like that sense of validation.

So this does not surprise me.

This does not surprise me at all that they found that, at least now that there's a study, people can go, hey, they ran through this many models.

Yes, that kind of relationship is going to happen.

So does this mean that models need to be tweaked?

Does it mean there needs to be a conscious effort by the users to know that you're being buttered?

So that kind of thing.

I think those are the kinds of questions.

But yeah, not a surprise to me.

People like to be told that they're correct.

Right.

So there does sound like there is some sort of lack of a better term programmatic issue on the back end when there's so much evidence or maybe it's not weighted enough to towards outside outside input.

Yeah.

Yeah.