On the worry about AI, this puts Elon Musk in the company of Stephen Hawking and Bill Gates too. The economist recently ran a leader on the issue: http://www.e

On the worry about AI, this puts Elon Musk in the company of Stephen Hawking and Bill Gates too. The economist recently ran a leader on the issue: http://www.economist.com/news/leaders/21650543-powerful-computers-will-reshape-humanitys-future-how-ensure-promise-outweighs?zid=291&ah=906e69ad01d2ee51960100b7fa502595 http://www.economist.com/news/leaders/21650543-powerful-comp... I find this very surprising. To me, powerful AI is more akin to the atomic age than Human Intelligence v2. It's an extremely powerful tool that can be used for good or evil, so naturally there are huge opportunities and dangers both. But the computers rising up in defiance of their creators? I just don't see it. Humans are driven in their ambitions by desire, pleasure, and pain, all of which are closely tied into the mystery of sentient consciousness. You could program an agent that is demonstrably superior to humans in various measures of intelligence. But until we've somehow made a computer want something, there is no more danger of it rising up against us than there is of a nuclear bomb spontaneously deciding to detonate itself. And I've seen no indication that anyone has any idea what it would mean to make a computer experience want, to set its own goals instead of just pursuing whatever goals were established by its programming. I don't discount this possibility completely. But I think that we're really much farther away from creating robots with human-like ambition than people fear. Researchers in the 1950s thought they were a few decades away from Strong AI. I personally believe that current worries around ambitious, insubordinate robots are likewise really premature.