Whenever AI is mentioned lots of people in the Linux space immediately react negatively.

Because whenever AI is mentioned it usually isn't even close to what AI meant.

The Linux community has never been of one mind on anything. We have always been against, and for, everything.

Some distro or project will integrate AI, or not, and it will be forked. And then forked again.

Many AI models are run on Linux. Linux won't be left behind in any real sense. Linux won't lose market share over this.

Linux developers paid by AI firms will integrate it into products. Those that volunteer will make their own decisions.

Imo you immensely overestimate the capabilities of these models. What they show to the public are always hand picked situations even if they say they dont

I'd call it realistic, not concerning.

i think firefox shows that ai can be used right, to help with accessibility.

i think the problem with ai is when companies use it as a buzzword instead of actual innovation by just cramming a bunch of ai into their product to do a bunch of niche things.

Using an Al-less desktop may be akin to hand copying books after the printing press revolution.

Or perhaps not.

AI is massively wasting power that we need for electrifying transportation and more useful things.

There are many things more useful than AI, for example good internet search engines.

AI can be useful for dedicated things like being trained on relevant tutorials and documentation to help with Linux.

yeah i see that too. it seems like mostly a reactionary viewpoint. the reaction is understandable to a point since a lot of the “AI” features are half baked and forced on the user. to that point i don’t think GNOME etc should be scrambling to add copies of these features.

what i would love to see is more engagement around additional pieces of software that are supplemental. for example, i would love if i could install a daemon that indexes my notes and allows me to do semantic search. or something similar with my images.

the problems with AI features aren't within the tech itself but in the surrounding politics. it’s become commonplace for “responsible” AI companies like OpenAI to not even produce papers around their tech (product announcement blogs that are vaguely scientific don’t count), much less source code, weights, and details on training data. and even when Meta releases their weights, they don’t specify their datasets. the rat race to see who can make a decent product with this amazing tech has made the whole industry a bunch of pearl clutching FOMO based tweakers. that sparks a comparison to blockchain, which is fair from the perspective of someone who hasn’t studied the tech or simply hasn’t seen a product that is relevant to them. but even those people will look at something fantastical like ChatGPT as if it’s pedestrian or unimpressive because when i asked it to write an implementation of the HTTP spec in the style of Fetty Wap it didn’t run perfectly the first time.

There is no ethical computing under capitalism. https://youtu.be/AaU6tI2pb3M?si=UfkoaSU-gTIvP52i

personally im fine with machine learning, what I don't like is "AI", a new marketing buzzword that justifies every shitty corporate exec decision and insane company evaluations.

There is a this app called upscale, that uses an ML model to upscale images. It's Quite good at what it does, it's useful. I use it frequently. So, there are AI stuff in linux. Just not your myopic view of AI (LLMs). And your analogy with printing press is extremely wrong. Other than human errors, printing press didn't have have remotely as many errors as LLMs. LLMs have not evolved to the point of causing a a revolution. So linux has plenty of time to see if the bandwagon sinks or sprints.

People who consciously use and support F/LOSS usually do it because they look at software with a very critical eye. They see the failures of proprietary software and choose to go the other way. That same critical view is why they are critical of most "AI" tools -- there have been numerous failures attributed to AI, and precious little value that isn't threatened by those failures.

I think most of the hostility is in regards to shilling of certain sites and services. Local self hosted AI is not likely to get as much flack I feel. Another aspect of hate is people generating images and calling it art, which...it is but, it's the microwave equivalent of art. Such negative sentiments can be remedied by actually doing artistic shit with whatever image they generate, like idk, put the image into Photoshop and maybe editing the image in a way that actually improves it, or using said image as a canvas to be added onto or some other shit.

Edit Addendum: also the negative perception of AI has mostly been engendered by some of its more unpleasant supporters, who think of it as a way to make "irrelevant" certain groups they don't like, and to take some sorta sick schadenfreude in the "replacement" of these people, which they think may be a way of reducing the power of these people (politically, socially, etc), and that's kinda fucked up.

Testing AI (knowledge system) was the first job out of college for me in the '90s (I used to be a programmer). I'm not against it, but I don't like it in my feet either. I like using the operating system all by myself, or generating things on my own. Especially now that I'm an artist, I like painting on paper. I even dislike digital art (I find it flat), let alone generative art.

As I mentioned in another comment we have an example of something like an ai-less desktop anology wise. gui-less installs. They are generally called server version of the distro and are used in datacenters but im 100% sure there are individuals out there running laptops with no gui. Im find with FOSS ai and there are LLM's licensed as such. That being said they are still problematic since the training requires large amounts of data that companies are not exactly strigent with collection.

General Ludd had some good points tho...

I'd argue that if you exactly call the model you refer to by their actual name, you'll get much different reactions. For instance, expert systems have been around for a long while.

I think conceptually AI is very useful and interesting, and as a general technical thing. But when we start talking about OpenAI and others, their methods for data collection, respect of licenses etc is where I (and I believe others) take issue

AI may be useful in some cases (ask Mozilla) but it is not like what you said in the middle part of your post. Seeing the vote rate makes me feel a tiny bit better about this situation.

That's easy, move over to Windows or Mac and enjoy. I'll stay in my dumb as Linux distros, thank you.

This article should be ignored.