While Corpos are pretty desperate for LLMs, putting enough roadblocks, costs, and other barriers in place will deter them. The profit seeking fools will chase a new god to bring them everlasting wealth, as they always have. Eh, China also makes unsafe models and ones that lie about confirmed points in history that China doesn't like to acknowledge because it goes against their narratives. Propaganda through tech, but to the LLM lovers, their models are efficient and do run on lower spec machines easier. The bubble will burst because they can't sustain the momentum and financial investment that they've poured into this. It isn't making them money, it costs them money. NVIDIA is the only one who has actually made any profit out of this, because they sell the hardware needed for Data Centers. NVIDIA then invests in the ChudAI companies to keep them floating just above the disaster line. However, the parasites are panicking because not enough people are paying the prices needed to keep this whole sham afloat. They are hemorrhaging capital at a rate which makes investors wary.

Yeah, I think LLMs should've been used for their original purpose: The processing of language, understanding how it is used, etc. The current usage that the corpos are using it for are...Disturbing and honestly aren't going to help humanity overall, its just a big grift and that makes me sad. They are trying to find a way to channel skill into their hands and deprive the skill of wealth, it's become a Torment Nexus of disenfranchisement, spying, misinformation, disinformation all in one. A very useful thing for those with ill intentions, malicious actors immediately leapt upon those things and wreak havoc with gassed up LLMs.

Linus, was naturally wrong, I think he is probably in the early stages of AI induced psychosis...It hasn't reduced his ability and technical skill yet. But I fear it will in time, the more he trusts gassed up LLM outputs. It already shows with his mandates for Kernel Development, it is open season for technofascists behind AI.

Personally, I am more of a troubleshooter, that does have a kind of ability with figuring stuff out (reading documentation + reading the error messages) who is also learning Python, I will learn C as well. I have been using Garuda Linux (Arch-based, BTW) it's pretty solid and wonderful, save for the bane of systemd. I can't say that it isn't problem free for me, as I have a lot of small but resolvable issues (often taking no more than an hour of reading and a few minutes of pinpointing and implementing the fix). However, immutable distros are fairly solid too, it is just a shame that Bazzite got compromised as well. But, it is what it is. As long as it doesn't completely become a vibe-coded mess like Bluefin is transitioning into, you'll probably be fine for the time being. For me, I love them because there were ways of doing the same things you could on regular distros, so it was as if you were just using a regular distro (but you aren't). I think from a security perspective, they are really good. But the container dependencies need to be unslopped before I decide to install one again, because I don't want to have any AI touched thing on my computer.

HTML is nerd sphere stuff, so you can't sell yourself short, it's just human readable like Python is (which Python can easily integrate with HTML for web stuff).

Yeah, when I am typing rarely do I actually feel pressed or angry, I am just trying to state my opinions in a way that someone else can understand. Thinking with clarity and peace is honestly helpful to this end.