A group of four female athletes are in various states of repose after completing a relay event at the Olympics. Three are looking toward the upper right at something offscreen; the woman on the left has anxious tears in her eyes. The fourth young woman is facing the woman on the left who is crying and comforting her. All are wearing orange racing tank tops and blue athletic shorts. They all have long hair that is either partially up and braided or completely braided.
Was this description written by an AI bot? Was this post?
Re: the rules on the side: “Alt-Text for vision-impaired users in the post body or in the comments is highly encouraged. Just pretend you are describing a photo to someone on the phone.”
Some instances have “alternate text” boxes that don’t show up for everyone but allow screen readers to read it, but if not, just put it in the body. Or as a comment.
It’s alternate text for people with visual impairments/screen readers.
I wrote it.
Maybe it’s written for people with visual impairments.
It would be pretty impressive if an AI could create a description that is as accurate as that. Obviously the description isn’t perfect but it’s not bad either.
That could help a lot with accessibility if AI-enhanced screen readers could interpret images too.
It would actually be interesting to use the descriptions here as a prompt for AI to see what it comes up with and compare.
https://sh.itjust.works/comment/1115896
I did try but it’s just too difficult for AI to accurately capture the emotional expressions of the athletes without further interference.