Emily Grundy comments on Why people want to work on AI safety (but don’t)