If it's meaning you want from art, then DALLE-3 or SDXL plus extensions or abilities like inpainting might actually be the most effective way to achieve that.

If it's meaning you want from art, then DALLE-3 or SDXL plus extensions or abilities like inpainting might actually be the most effective way to achieve that. After all, there is quite a bit of meaning in the text used to generate them. That seems like a head start if you are trying to make a statement. Not to mention the speed that the image or image elements emerge and can be iterated or integrated. For example, take Judy Chicago's The Dinner Party (admittedly just from the top of Google search results about artistic statements). It is a series of plates that have sculptures depicting femininity in various ways at various heights. With something like a painting tool with SDXL or DALLE-3 integrated one could create an image with similar content by drawing the table and then generating/inserting or inpainting each of the individual plate contents per specification such as the size of the inpaint mask and prompt. She clearly explained the meaning of the work and it can fit in a few long sentences. The more I think about this, the more it seems that image generation should be very helpful for artists who want to "say" something.