Yudkowsky has a a handicap too, since presumably he feels that if he says anything too insightful, particularly concerning AI architecture, there’s a small probability that insight might be used to accelerate Unfriendly AI development.
Yudkowsky has a a handicap too, since presumably he feels that if he says anything too insightful, particularly concerning AI architecture, there’s a small probability that insight might be used to accelerate Unfriendly AI development.