Yes – definitely giving samples and Prompting AI in front of them and showing the results then critiquing it for the genericness? But reprompting multiple times can produce better results … but it also means we are doing more thinking and writing ourselves. What do you think?