maybe a solution would be if the ai can provide context. e.g. here is an image of a nigerian person with certain characteristics, but the possible result space

maybe a solution would be if the ai can provide context. e.g. here is an image of a nigerian person with certain characteristics, but the possible result space is bigger and here you can try with nearby prompts which will return different results.