3 ms·

This isn't what "unreasonable effectiveness" is. Why worry about AI beliefs being disconnected from source meaning. Humans already abuse language.