Thanks for the response. My hope for AI Alignment studies is that it would be something rigorous and theoretical and be testable in current AI safety issues. Mu

Thanks for the response. My hope for AI Alignment studies is that it would be something rigorous and theoretical and be testable in current AI safety issues. Much like studies of abstract algorithms can be compared and contrast with the runtime of actual algorithms. What I have seen in LessWrong is lots extrapolations upon extrapolations with worse case scenarios just to say there is nothing we can do. That seems to take away from any serious research that might exist since they are probably the best known community associated with AI Alignment.