Shmi comments on Some abstract, non-technical reasons to be non-maximally-pessimistic about AI alignment

[Shmi]

How many of those results are accepted as interesting and insighful outside MIRI?