What's to miss out on? Experimenting with or even building useful "AI" tools? Nope. Not missing out on that. There's Ollama (and a bunch of other) local (yo

What's to miss out on? Experimenting with or even building useful "AI" tools? Nope. Not missing out on that. There's Ollama (and a bunch of other) local (your machine, not someone else's) "AI" model runners native to Linux. There's also a ton of great Python libraries which connect just as happily to a local "AI" instance as they will to a remote API. Much "AI" research already happens on Linux. Using other useful tools other people have written? Nope. Not missing out there either. Same reason. Tons of great tools already exist to run "AI" assisted tasks of various kinds natively on Linux connecting to a locally running model. All that stays securely on my machine / network, and doesn't "phone home" (as can be confirmed by reading the source code of the open source "AI" tools by anyone with a bit of programming skill / knowledge). The difference is that I control when and how it runs, what it does, where it gets data from, where that data flows to, etc, etc. Not Microsoft or Apple, nor anyone else. Me. That's about the only way I can trust current "AI" to be used properly and safely.