[Link] Humanity in Jeopardy - Max Tegmark on AI risk — LessWrong