I envy computers. They don't stress over false beliefs and processes. Sure, bugs in software will make a computer produce results that are not very useful, but they don't suffer from self-criticism in the face of useless results. When we humans make a mistake, it is worse than a useless result; it can be experienced as shameful failure.
On the other hand, without shame or guilt, a computer has no impetus to change itself and no criterion to judge whether a change is in the right direction. But who judges the judge? How do we know shame is an appropriate motivator?