I was terrible at this task as a kid. When asked to describe an image I usually volunteered single piece of information about it and I had to be prompted multip

I was terrible at this task as a kid. When asked to describe an image I usually volunteered single piece of information about it and I had to be prompted multiple times and asked leading questions to observe and describe more of it. GPT-4V beats me even now. There were a lot of information in the descriptions I wouldn't notice or include without being specifically asked about them.