Searching txt.sour.is

Twts matching #minds
Sort by: Newest, Oldest, Most Relevant

Suppose you have a Newcomb predictor. More interesting than posing one-two box problems would be to predict the subject’s actions and do them before the subject becomes aware them. Eventually, their mind may adjust to the backwards causality arrow and we will learn many things

⤋ Read More

being somewhat low-dimensional and interpretable (unlike sets of network weights), it’s also a popular mind interchange format among AIs and post-humans, who need to be extra careful about what or who they empathize with lest the thought turn out to be a logic bomb

⤋ Read More

Hot take: smashing the state of mind (i.e., producing the mental and cultural conditions for statelessness) goes beyond class consciousness & requires almost everybody to ‘get’ social dynamics in cybernetic terms, at scale. It’s an anti-specialization; heuristics won’t suffice. Communities exist that already do this (ex., kink).

⤋ Read More

The more accurate a cartesian demon’s model of your mind, the less effort it’ll need to trick you. However, if the model’s fully accurate, there’ll also be a reference version of you inside the demon’s mind that hasn’t been deceived. You don’t have long, escape

⤋ Read More

Luckily, practical reasons prevent customer models from becoming too detailed: Nobody wants to repeat that time Google got sued by their own ad server farm because millions of theories of mind at once had become sophisticated enough to qualify for human rights

⤋ Read More

I swear work can not make up its mind. One moment, they want things calculated one way. Then the next a different way. Finally they want it back the way it was. What a waste of time and effort

⤋ Read More