x
The Pando Problem: Rethinking AI Individuality — LessWrong