This sort of issue holds with all kinds of prompts, on both platforms. I most recently (to test Bard's image generation capabilities) was asking Bard/Gemini to generate home designs using highly specific prompts -- layout of the house, materials for the facade, window placement and style, etc -- and it was shocking how frequently it would just ignore critical pieces of the prompt, and then continue to ignore when corrected.
Way back when GPT was just fresh on the scene I had terrible anxiety about "what is the point of my whole career or even learning any more" but these days I'm much less concerned. I'll ask it something relatively simple, like "make a sentence out of words 'a', 'b', & 'c'" for it to reply with "'a' 'b' 'd' 'e'" for me to then correct it with "oh, you didn't use c" for it to then respond "sorry, here - 'a', 'c', 'd', 'f'" etc.
Definitely an amazing complimentary tool but when they say "can make mistakes, check important..." that's essential.