The should-universe is the imaginary universe in which everything works the way it common-sensically ought to, as opposed to the actual is-universe we live in.

Eliezer Yudkowsky discussed "living in the should-universe" as a common failure mode for wannabe AGI programmers in Above-Average AI Scientists, and illustrated two examples of how living in the should-universe can be dangerous for an AGI programmer: (1) being delusionally optimistic, and (2) treating the should-universe as the point of departure — describing the real universe as the should-universe plus a diff.

Living in the should-universe is a general problem that is applicable to more than just AGI programming, and there are many other ways in which living in the should-universe can lead to error than just the examples given above.

Posts

Wikitag Contributions

Comments