As a contractor who built a lot of predictive systems and workflows in last three years I can tell you that quite often there is a specific request to put AI into it even when it is not needed and would objectively make the system worse, slower and more expensive.
The AI psychosis is a real thing.
I keep seeing requests to replace what would be a perfect UNIX shell script with agents, like what is the benefit other than being able to say we're doing AI?
Where I work, management hasn't considered integrating AI at all, yet some clients are very vocal about it being the future and worry we are going to be left behind. Most people just don't care, and I worry the squeaky wheel will eventually get the grease.
So then, do you put AI into it anyway because they asked for it, or do you tell them that you won’t do that?
Maybe it should have clicked earlier in life and I'm perhaps that much dumb dumb, but it only recently occurred to me (from experiencing it at two very different companies and discussing with peers having reached a certain seniority level more or less at the same time) how dysfunctional many companies are, and how often they produce incentives that are misaligned with the overall company goals and sustainability principles. I blame in large part a layer of middle management that selfishly puts itself above all else, misguides, misrepresents, because it essentially pays larger dividends (literally and not) to "play the networking game than to be an efficient and effective productive structure". Maybe that's to be expected in a services-driven economy where the value of the work is immaterial and subjective (and the whole phenomenon of bullshit jobs).
Haha, i have a colleague, he is the "AI-is-for-everything-let-me-check-Claude-first":
Regardless which task is handed to him, he "discusses" it first with Claude and very often comes back with like "The AI said... X"