But is there a particular area where AI causes hesitation or adds bad friction? And I know there's good friction and there's bad friction, but what are those differences? Yeah,

absolutely.

So again, I don't think it's... one or the other, AI is good or AI is bad.

Like there's nuance where it's typically needs to be applied a certain way because again, the job in healthcare is a lot more high risk.

It requires a lot of judgment.

And if AI is applied poorly, it can add more confusion and more frustration and can create uncertainty if it's not implemented well.

So from a UX perspective and behavior design, the things that we see often is, If AI is not applied well, that can increase hesitation, as you said.

What would you say people should do to keep that up? What kinds of things can we actually do to be proactive? So I have a couple of