x
The Case for Artificial Manifold Intelligence — LessWrong