The prompt "A ten year old boy daydreaming" is a bit haunted, as it's not specific enough to generate good results. But the results from running that does not

The prompt "A ten year old boy daydreaming" is a bit haunted, as it's not specific enough to generate good results. But the results from running that does not result in NSFW content, contrary to what you say. Most of these NSFW "blockers" simply act on keywords like "ten year old boy" rather than trying to understand the output. Here is 9 images of the prompt "A ten year old boy daydreaming" with various settings. None of them are good (bad prompt + no tuning on my side), but none of them are NSFW either: https://imgur.com/a/FcBv05w https://imgur.com/a/FcBv05w (slightly off-topic, seems there is a lot of misinformation in this thread, from misunderstanding the training data, to how the NSFW content block some platforms are implementing works)