I think that AI is inevitable, but I think that unfriendly AI is more likely than friendly AI. This is just from my experience in developing software even in my small team environment where there are less human egos and tribalism/signaling to deal with. Something that you hadn't thought of is always going to happen and a bug will be perpetuated throughout the lifecycle of your software. With AI, who knows what implications these bugs will have.
Rationality itself has to become much more mainstream before tackling AI responsibly.
A long blog post explains why the author, a feminist, is not comfortable with the rationalist community despite thinking it is "super cool and interesting". It's directed specifically at Yvain, but it's probably general enough to be of some interest here.
http://apophemi.wordpress.com/2014/01/04/why-im-not-on-the-rationalist-masterlist/
I'm not sure if I can summarize this fairly but the main thrust seems to be that we are overly willing to entertain offensive/taboo/hurtful ideas and this drives off many types of people. Here's a quote:
The author perceives a link between LW type open discourse and danger to minority groups. I'm not sure whether that's true or not. Take race. Many LWers are willing to entertain ideas about the existence and possible importance of average group differences in psychological traits. So, maybe LWers are racists. But they're racists who continually obsess over optimizing their philanthropic contributions to African charities. So, maybe not racists in a dangerous way?
An overly rosy view, perhaps, and I don't want to deny the reality of the blogger's experience. Clearly, the person is intelligent and attracted to some aspects of LW discourse while turned off by other aspects.