Vibe-Coders, people that use LLMs for programming, are found at a such high levels of stress these days, that they [end up in a hospital](https://invidio.us/watch?v=0F5zG1RQ4Ig) over it. They believe that they need to keep up-to-date on everything regarding the endless AI updates, to the point that it literally deteriorates their health. We all know a person in some field that feels like they would not necessarily *want* to use AI, but in the same time, they *don't want* to "fall behind" the rest. That they must know how to "use" these tools, or else they will have no chance what so ever, in the market-place. The problem is, they are focusing on the wrong Fear. Especially when it comes to the market-place.

It is quite ironic that "vibe-coders" people who don't have to *think* when developing software, have such a high level of anxiety, that it affects their health. AI's whole selling point is that you don't have to do anything. That you can fire your work-force and let the algorithm do all of the work for you. AI is specifically designed to be useful to mindless managers that don't know nothing. 

Yet it feels like actual programmers, in a desperate attempt to save their job positions, or something, cling to this algorithm, like a swarm of bees. An algorithm that is designed to replace the actual programmers.

But I'm not talking about how AI replaces the workforce. I'm talking about something much weirder.

We already saw companies re-hire some of their work-force after trying to replace them all with AI chat-bots. AI fan-boys would proclaim, that it was because the AI models themselves weren't ready yet. And soon, when they are ready, nobody will re-hire nobody. It will be all just AI. But I believe it's not so simple. And soon companies will need people specifically who never used AI in the first place.

Let me attempt to explain it.

GNU / Linux operating system seen a monumental growth since roughly the time of the launch of Microsoft Co-Pilot ( a year earlier than ChatGPT ).

Now I know about the whole Linus Torvalds remark about vibe-coding and how he is not against it, or whatever. But knowing Linus's previous shenanigans, you can safely assume, he is just trying to soothe the corporate sponsors of his operation.

Putting this aside, the growth of GNU / Linux can be attributed directly to the in-house use of AI coding agents inside of Microsoft. Making Microsoft Windows ( a place that loses it's market-share to GNU / Linux ) a pile of absolute, buggy garbage.

Let's trace it back:

 - 2005. Linus Torvalds writes GIT. A version control software specifically to manage the development of Linux. ( The kernel )

 - 2008. Github.com opens up. A web-interface for GIT.

 - From 2008 to 2018. People that want to develop any sort of Libre Software, usually host their source code on Github. Minus some special cases that host elsewhere, or on their own website.

 - 2018. Microsoft buys Github. ( Prompting people to look for an alternative hosting platform )

 - 2021. Microsoft releases GitHub Copilot. An AI coding tool trained on the source code hosted on Github.

 - 2021 - 2026. Microsoft uses coding Agents ( copilot and others ) on Microsoft Windows. Making the code-base of the operating system increasingly more ridiculous. Driving the user-base to switch to GNU / Linux.

*Some might say, this was Linus's plan all along.*

People ran away from Windows because of vibe-coding. Because of the bugs it introduced in the system. Similarly to how people ran away from Github. Today a large majority of Libre Software is hosted on Codeberg.org instead. Similarly to how a lot of people ran away from Ex-Twitter, when Musk bought it. Before that Mastodon was a desert. Now it swarms with life.

Today I came across an article about [some stupid AI short-film, by a big name director](https://aftermath.site/neill-blomkamp-ai-movie/) and there was this quote:

> He is terrible at telling stories because he doesn't *want* to tell stories.

He doesn't *want* to tell stories... People that use AI don't *want* to do *something good*. They don't want to create something worth your time. They maybe want money. And they realized ( using analytical methods or something ) that you will not care, unless something looks a certain way, or behaves a certain way.

Managers understand that paying for people who care to make something good, is, first and foremost, an expense. And expense that will be subtracted from the gains, in the annual report. So they are desperately trying to come up with a "solution", to cut that expense, while still gaining all this gain.

Think about it. Say movies. For the longest time, impressive big budget block-busters with great visuals and impressive everything, was the shit people wanted to see. So if a manager can find a way to make something that looks as good and feels as good, while being a thousand times cheaper to produces, that manager will be a good manager. AI makes perfect sense in this setting.

But the manager doesn't realize something very crucial about the whole ordeal. We are not watching these movies because they look the way they look. But instead we infer that they are good movies from the way they look. Good movies aren't pretty movies. Good movies are movies that are made by people that cared. It just so happened that for the longest time only people who cared made pretty movies.

Same goes for everything else. We want our stuff to be made by those who care. We want software that is programmed by programmers that want it to be useful for us. We want appliances to be designed by designers that think of the user. We want things we read and music we listen to and pictures we look at to be made by an artist who cares.

By definition, an AI viber, by prompting an algorithm to generate something, shows that they don't care. So why should we?

When the hype dries out, when the bubble collapses, those of us that swore never to touch the damn algorithms will be in the high demand. Because we were the people who the users never doubted. Never doubted that we cared.

**Happy Hacking!!!**