I'm writing this from a crappy laptop with 2GB of RAM and a dull screen.

But how would you implement that new Microsoft Screenshot surveillance bullshit feature? Just imagine what a giant waste of resources that is. You have something on your screen which is information and mostly likely already in a good form to process like text. But it makes a screenshot every few seconds and uses some "AI" to make the already existing information searchable again from a fucking screenshot??? Maybe I missed something but that is how I understood the feature.