The driving force behind rationalism, and founder of the Machine Intelligence Research Institute, seems to think it's already too late for humanity; we are already doomed to a catastrophic AI apocalypse. How worried should we be?