I am sure it can change. Ultimately programming we be completely lost; we are going to hand-wave and command computers inside us or inside the walls of our houses. This absolutely will happen.

But my issue is with people that claim we are either at that point, or very nearly it.

...No, we are not. We are not even 10% there. I am certain LLMs will be a crucial part of such a general AI one day but it's quite funny how people mistake the brain's frontal lobe with the entire brain. :)

Plenty of jobs for coming by to read the logs across 6 systems when the LLM applications break and can't fix themselves.

Yep, quite correct.

I am even fixing one small app currently that the business owner generated with ChatGPT. So this entire discussion here is doubly amusing for me.

Just like we were fixing last year the PowerApps-generated apps (or any other low/no code app) the citizens developers slapped together.

The question is how those jobs will pay. That seems like something that might not be able to demand a living wage.

If everyone prompts code. Less people will actually know what they're doing?

That's the part many LLM proponents don't get it or hand-wave away. For now LLMs can produce okay one-off / throwaway code by having been fed with StackOverflow and Reddit. What happens 5 years down the road when half the programmers are actually prompters?

I'll stick to my "old-school" programming. I seem to have a very wealthy near-retirement period in front of me. I'll make gobs of money just by not having forgotten how to do programming.

Really? Excellent debugging is something I associate with higher-paid engineers.

If it can't demand a living wage then senior programmers will not be doing it, leading to this software remaining defective. What _that_ will lead to we have no idea because we don't know what kind of software.