I agree with the mix of it being shops that simply aren't moving on it as fast and shops that just assume it as part of the workflow without needing to list it as a skill. It's not like the barrier of entry to using AI in some coding capacity is that low, which is why whatever you consider a senior-level engineer still makes the difference when the barrier of entry is arguably lower and the ability to express intent, precision, and perform good review is valuable and comes from experience of some kind.

And I think we can face that there are people that may be "behind", but there also are always plenty of businesses that can stay in some sort of stone age, because they don't need cutting edge SaaS etc. to continue making a profit. Some profit-making non-tech companies just have the tech they need and don't feel the need to hyper optimize it. Varying per location and industry, you never know what tech is still being maintained and/or how much potential red tape around employee tooling there is.

Those types of jobs aren't necessarily engineering-ly rewarding if you're a coding goblin fiend like me, though if I was paid well to do AI-less work for someone, I'm not sure if I'd actually care that much. Typing is typing, I know my main domains fairly well from AI-less experience, I'm over going above and beyond what's expected of me as long as I feel like I can employ design that doesn't drive me insane, and I can do what I want in my free time.

Some of those jobs can be places of stagnation, but some tech teams can really survive without changing like roaches, though members can always try to improve their skills and such outside of those constraints when they hopefully can. I don't think that's really changed with any of recent previous generations of most forms of technology in general. No matter how fast tech moving, there is still always a large momentum to old ways, and perhaps adoption windows of new large patterns don't really scale in the same way or rate as the technology itself.