[link] Essay on AI Safety

I re­cently wrote an es­say about AI risk, tar­geted at other aca­demics:

Long-Term and Short-Term Challenges to En­sur­ing the Safety of AI Systems

I think it might be in­ter­est­ing to some of you, so I am shar­ing it here. I would ap­pre­ci­ate any feed­back any of you have, es­pe­cially from oth­ers who do AI /​ ma­chine learn­ing re­search.