Comment by ziml77

No it just seems that it becomes blind, so to speak, to the negatives and the inclusion of the words you were negating makes it more likely to apply them in the positive. This is how ChatGPT has seemed to behave whenever I've tried to get it to not include something.