Automating a broken process just speeds up the problem. AI vendors oversell capabilities, but garbage data and flawed processes still produce garbage results. Fix your data and process first, or AI will fail.
I really found myself nodding my head with Anirban's comment there about, you know, how if you're automating a broken process, you're just accelerating the problem. And I would say that that can happen when, especially in this day and age with AI, there tends to be a belief that AI is magic. Either, you know, we're kind of our worst enemy, some of the AI vendors, because we like, you know, AI is an attractive thing, Yeah. People like to hear about it, but we have to be careful not to oversell it because it's not going to work when you're— when the data coming in is, you know, invalid or problematic or the process itself is wrong. So, you know, the classic garbage in, garbage out.
