Luke Muehlhauser’s views on AI safety

http://lesswrong.com/lw/iqi/intelligence_amplification_and_friendly_ai/

http://lesswrong.com/lw/i0a/recent_miri_workshop_results/9g87

http://lesswrong.com/lw/ffh/how_can_i_reduce_existential_risk_from_ai/

http://lesswrong.com/lw/e97/stupid_questions_open_thread_round_4/7bg7

http://lesswrong.com/lw/e97/stupid_questions_open_thread_round_4/7be1

http://lesswrong.com/lw/cua/strategic_research_on_ai_risk/

http://lesswrong.com/lw/bjl/ai_risk_opportunity_strategic_analysis_via/

http://lesswrong.com/lw/bdt/ai_risk_opportunity_questions_we_want_answered/

http://lesswrong.com/lw/9oq/link_2011_team_may_be_chosen_to_receive_14/5zox

http://lesswrong.com/lw/ajm/ai_risk_and_opportunity_a_strategic_analysis/

http://lesswrong.com/lw/ah5/singularity_summit_2011_workshop_report/5xyy (and also see other comments on that same post)

http://lesswrong.com/lw/91c/so_you_want_to_save_the_world/ (this one is marked as “very out-of-date”, but I think many of the questions are still relevant)

http://lesswrong.com/lw/8ts/open_problems_related_to_the_singularity_draft_1/

http://lesswrong.com/lw/91c/so_you_want_to_save_the_world/5jdo