< the skull

AI IS NOT COMING. IT’S ALREADY SITTING AT YOUR DESK
===================================================
2026-04-30

Not liking artificial intelligence in today’s world is starting to feel a bit
like being the grumpy old man yelling at clouds about “the good old days.”

It is like swimming against a tide that has absolutely no intention of turning
back. Like guarding some imaginary national border and shouting, “No foreigners
allowed!” at a technology that already moved into your house, connected to your
Wi-Fi, and started helping your cousin build an app.

In other words: it is not just pointless. It is probably wrong.

AI is here to change the game. Get used to it, welcome it, and, more
importantly, use it. There is no realistic way to stop this kind of progress.
You can complain, you can resist, you can write very angry Reddit comments, but
the train has already left the station — and it is generating Python scripts on
the way.

Lately, I have seen a lot of rejection from the developer community. IRC
channels, Reddit threads, comment sections — the usual places where hope goes to
die. And the main card people keep playing is the security card.

“AI-generated code is unsafe.”

Sure. Sometimes it is.

But let’s not pretend software was a perfectly secure paradise before AI showed
up wearing sunglasses and holding a JavaScript framework. Security flaws existed
before AI, and they will continue to exist after it. Bad code was already being
written by humans with confidence, caffeine, and terrible variable names.

The idea of “not wanting AI-written code” is, at this point, more of a wish than
a plan. Because it is already happening. People who barely knew how to turn on a
computer can now build systems that, in some cases, are surprisingly complex.
Are those systems always good? No. Are they always safe? Also no. But pretending
this shift can be stopped is like trying to uninstall the internet.

What I see for the future of programmers is a change in role. Developers may
become less like bricklayers and more like project directors. Less typing every
single line by hand, more reviewing, guiding, testing, supervising, and making
decisions.

And honestly? That is not a bad thing.

I like using AI to code. I have learned a lot from it. I have also made a lot of
mistakes with it. And no, I do not think it should always be the only option.
There were plenty of times when AI sent me in circles around a simple problem,
like a very confident GPS telling me to drive into a lake.

Sometimes I knew the solution, doubted myself, trusted the AI, wasted time, and
then watched the AI politely agree that I had been right all along.

Beautiful. Annoying, but beautiful.

That is the point: AI is useful, but it is not magic. It is powerful, but it
still needs supervision. It can speed things up, teach us, help us explore
ideas, and make programming more accessible. But it can also hallucinate,
overcomplicate, and produce nonsense with the confidence of a senior developer
in a meeting.

So maybe the answer is not to fight it.

Maybe the answer is to supervise it, learn from it, question it, and take
advantage of what it does well.

Because as intelligent as these systems seem — or almost seem — the people
building them, using them, correcting them, and being responsible for the final
result are still human.

At least for now.


< the skull