Whenever AI is mentioned lots of people in the Linux space immediately react negatively. Creators like TheLinuxExperiment on YouTube always feel the need to add a disclaimer that “some people think AI is problematic” or something along those lines if an AI topic is discussed. I get that AI has many problems but at the same time the potential it has is immense, especially as an assistant on personal computers (just look at what “Apple Intelligence” seems to be capable of.) Gnome and other desktops need to start working on integrating FOSS AI models so that we don’t become obsolete. Using an AI-less desktop may be akin to hand copying books after the printing press revolution. If you think of specific problems it is better to point them out and try think of solutions, not reject the technology as a whole.
TLDR: A lot of ludite sentiments around AI in Linux community.
I imagine it might happen one day. But at present, I don’t really think that most computers are at a point where they can utilize it without the use of proprietary cloud technologies that aren’t considered to be ethical nor financially sustainable. And even if people’s computers could fully handle things themselves, there would still need to be a group of developers with enough knowledge to actually implement it.
Consumer AI has always been pretty limited in most Linux desktops. Heck, I’m still waiting for a Desktop Environment to one day have a nice implementation of Speech-to-text like Windows and macOS.
I feel like another part of it too is just that Linux users also just have higher expectations in areas around privacy, security, and flexibility, and lower expectations of elements like UX and Minimum Viable Product, the latter especially being that they don’t even view the software as a “product”.
A lot of AI features are powered by data collection in some way. And given that most Linux users don’t even like small amounts of telemetry being sent without their explicit permission, I couldn’t imagine how libre AI models could be built, especially on a shoestring budget, to produce something that would be capable of producing acceptable results. All without avoiding the heat that current AI companies are facing with plagiarism accusations and copyright infringement.
I’m not really saying it can’t happen, But it would require a larger organization like Mozilla, who’s actively working on building open source AI that could then be later incorporated by someone else (similar to the soon to be dead Mozilla location services being integrated through daemons used by desktop environments). Or, by a much more random guess, by a corporation with a profit incentive to incorporate Linux like Valve and the Steam Deck with its inclusion of the plasma Desktop via an Arch fork. And in the long run, the FOSS community building a larger developer base that actually could, And one day upstream it all once it’s in a good enough format.