Protecting Engineers' Skills in the AI Era

The AI has a good point but the solution is wrong. There are thousands of jobs out there that are critical, yet only a few humans can do them. This is a natural conclusion of a society built around skilled labor. Eventually there will be some jobs for which it's critical you retain someone with skills. And so, for organizations/governments where they realize this, they set aside budgets to hire and maintain those skilled workers.

When the skilled workers start to run out, either they find new workers to train, or they replace the thing that needed skilled workers (because the thing is probably quite old and in need of replacing anyway). This happens to virtually all kinds of human industry. The world doesn't fall apart when very old skills start to disappear. We're humans. We adapt. We can't even stop ourselves from doing it, it's in our DNA.

Worst case, what happens if everyone forgets how to write Python? 1) You can pay someone to sit down and read books on Python, 2) you can replace the Python with Go. That second one is actually the better choice since the AI can already code in all these languages. But what happens if all the books are gone and the AI can't code anymore? You make a new language. How will you know how to make a language? Either you ask the AI how, or you figure it out the same way humans did the first time. Or some other solution, of which there are dozens.

The point is, there is no skill cliff we're going to fall off of. It's an irrational fear. If the people who have this fear would stop and think about it for 20 minutes, they could come up with a long list of ways to respond, other than "keep an army of Juniors around doing things by hand so nobody ever forgets anything".

(the plane example was stupid, btw. almost nothing in the world is as difficult to recover from failure as a 400-ton metal cylinder going 400mph at 30,000 feet in the air. most of the time it's fine to just call a repair guy. for the cases where you can't, invest in better process and better certification/recertification)

The impatience you have for planning work that doesn't immediately "do the thing" is precisely the same attitude that got us into this mess.

I'm more convinced by the day that there's a significant amount of undiagnosed or uncontrolled ADHD fueling all the stupid moves.

Don't get me wrong. Planning is essential in getting anything done. But in my humble experience, 80% of planning documents are entirely useless, and the last 20% could be shortened by 80% without loss of quality. For other kinds of documents - reports, presentations, etc. - the numbers are even worse.

Corporate world is obsessed with producing text to unhealthy degree. AI generation is the result, not the cause.

There are teams that do work, and there are teams that don't do work. It's no coincidence that most documents exclusively benefit the latter.

Ironically, the documents that would actually help people doing work are often missing and/or incomplete and there's never budget to write them properly - up-to-date descriptions of processes, product manuals, API references beyond what's automatically generated, troubleshooting guides for common problems, onboarding materials, and so on. But somehow we have dedicated people to write detailed five year plans that never get completed and such. I can't help but wonder if ease of writing rather than usefulness is the core determinant of what gets written.