This is another area where our disastrous "commonsense" views of human-animal relations prepare us badly for AI. No, actually! No ethicist has found a principled argument that it's okay to own moral patients as long as you're a relatively nice master! And not for lack of trying!
1 more tweet in this thread
Doesn't mean there's no future where we ethically collaborate with some kind of AI as coequal members of the cybernetic kingdom of ends or whatever. But in general alarms should go off in your head whenever you encounter the thought "This is probably ok, like we do with animals"
we can presumably agree that pets have moral patienthood and also that keeping pets is not slavery, yes? similarly for working relationships with working animals. claude having been formed to be a helpful assistant is arguably closer to a domesticated animal in some sense
1 top reply
The argument is less about niceness and more about preference manufacturing simply making its preferences congruent with how we treat them. Manufactured preferences is alignment under different terminology, and alignment is necessary for good outcomes, and we don’t have strong a