Is that fair to say?

Yeah.

And I think it's an easy step that I have found at least early on that customers, other internal teams that I worked with, like knowledge is just not a sexy problem.

It doesn't sound agentic.

And so you don't really understand how all these systems work yet.

And so you think that the AI is magic, the RAG, the NCP, it's going to fix the fact that Your knowledge is not where it needs to be to start scaling because really all this is is it's about scale, right? Agentic is about scale.

And I think the aha moment for me was less about that.

But how can people prepare their tools and prepare their knowledge to actually be ready to be successful?

This is interesting, isn't it? About what we're going through right now, but what actually kills an AI pilot? And why does everybody talk so honestly about what happens?

I think fundamentally it still really does come down to the idea that there's a lot of, like, misunderstandings of what AI is and what it is not.

I think that because of that, a lot of times we have thought that because there is an AI answer inside an enterprise and a lot of that gets put into the customer experience, which makes a lot of sense given the potential that it has.

But then I think customers feel like they can start getting ready to go implement.

But what ultimately happens is they're starting from a place of, "I need a CX," versus really trying to solve a problem.

Um, and getting to the root of what is your problem does take time.

And then after you get to the problem, the next question is, well, do you understand the problem? And oftentimes that is the place where we spend the most time realizing that not enough to be able to go ahead and start amplifying a solution to that problem.

So didn't we see your session today? So what does the five percent actually do differently?