Each model is created from its own unique dataset, which contains rules around the use of certain terms in prompts.

It does not necessarily mean that an illegal or rude term has been used, it is simply that a particular term is rejected by that dataset and therefore may not be able to generate the desired image.

If a 'profanity' message appears, it is useful to try and tweak the terms the prompt given to avoid specific terms that are rejected by a model, or switch to a different model available.