Because getting truthful answers would make users happier? Seems like common sense to me. Who’s asking the chat bot questions not looking for or wanting a tru

Because getting truthful answers would make users happier? Seems like common sense to me. Who’s asking the chat bot questions not looking for or wanting a truthful answer a lot of the time? If the model understood or captured “human interest” at all in its training this should be pretty fundamental to its behavior.