Mint.
I’ll be very honest with you. It’s not fancy, it’s not snazzy computing. It’s simple, designed with a graphical interface in mind, and a good operating system for someone who A) does not know Linux, or B) does not want to fiddle.
Congratulations. Your a system admin. For real.
I’ve interviewed candidates for system admin jobs who had less exposure to managing Linux then this story.
Not dissimilar - my three steps.
I’m of the mind that it can involve a screen - but needs to be a different part of your brain. For example 3d printing, writing, reading a (digital) book. However if you are nearing burnout, you need to pick up something radically increase your non work time, and spend your free time doing something that does not require a screen.
Full disclosure - my background is in operations (think IT) not AI research. So some of this might be wrong.
What’s marketed as AI is something called a large language model. This distinction is important because AI implies intelligence - where as a LLM is something else. At a high level LLMs are using something called “tokens” to break apart natural language into elements that a machine can understand, and then recombining those tokens to “create” something new. When a LLM is creating output it does not know what it is saying - it knows what token statistically comes after the token(s) it has generated already.
So to answer your question. An AI can hallucinate because it does not know the answer - its using advanced math to know that the period goes at the end of the sentence. and not in the middle.