Uh, so as AI become more and more embedded into workflows, how do organizations actually move, say, beyond isolated AI initiatives to build enterprise-wide orchestration capabilities that really create business value?

I think the pilot problem or the pilot issue is a real issue.

Uh, every organization that I know of has lots of proof of concepts and pilots.

Most of them don't scale, and the issue usually is not technology, it is about the operating model around the technology.

So I look at AI maturity the same way as the evolution of a GBS across the-- over the years, right? Where it moves from, uh, activity to process to outcomes, right? And in HFS own words, to a generative business services, right? So in AI also, I see the same evolution where you automate tasks first, then you move to a process, end-to-end process orchestration, and then you move to an autonomous AI agent-based workflow where it, it is orchestrating outcomes across the entire organization and across functions.

The-- In our experience in deploying AI at scale across functions, what we found is that organizations don't move straight to a, to a agentic operating model.

You first start with a deterministic model where you build governance, you build trust, you build reliability before you move to reusable agents.

So how does the operating model need to change when success is measured not by process KPIs, but by business outcomes? Any examples you want to share would also be useful.