What about at Duke? Out of all the technology, what's had the biggest impact, and how does AI fit into that?

Yes, I would say over the last five years or so, Duke was pretty heavily invested in doing GIS corrections.

So we had a fairly large scale project effort where we were, you know, collecting bulk data sets of, you know, distribution pole information and going in and doing GIS correction updates for ADMS.

You know, I think the real first kind of scaled AI implementation that we had was just we were getting all this data back and we were paying retired or light duty linemen to manually sift through thousands of images of distribution poles and identify data.

okay, here's a poll, the attribute says X when it's really Y, and we need to make that change in the GIS system.

And so one of the first things that they did was kind of start to use AI and ML a little bit to help those individuals identify, okay, here's what I'm looking at, you know, colored boxes and, you know, basic tagging and classification.

And so that's where we've been, I'd say, in the last five years or so, and we're actually getting ready to restart that at a limited scale again this year.

Awesome, I hope you get there.