For instance, a weak AI might refrain from visibly trying to produce smiles in disturbing ways as part of an effort (including verbal claims) to convince the programmers that it had apprehended the objective morality behind their attempts to inculcate smiles as a reinforcer.
For instance, a weak AI might refrain from visibly trying to produce smiles in disturbing ways as part of an effort (including verbal claims) to convince the programmers that it had apprehended the objective morality behind their attempts to inculcate smiles as a reinforcer.