Regarding that last example: the important part is, it does not look the same, nor does it feel the same, since the image is not developed under that regime of

Regarding that last example: the important part is, it does not look the same, nor does it feel the same, since the image is not developed under that regime of light and emphasis and attention. (In a way, there is neither a sense to light, nor detail, nor attention, and, thus, to "how?". I may add, if I'm engaging with an image, beyond a first impression, "how" is to me more important than "what".) Similar aspects may be observed with text. And, for you, as a writer: As you can't compete with generators on economic grounds, the productions of the former will probably become prevalent (if they are not already, in some niches). If input rejection becomes the default mode of reception, this may affect you, as well. PS: In other word, a representation is not the real thing, it's a thing of its own in its own realm, raising the question of "what realm?". As an illustration, here are two illustrations of my own, both representing real objects, drawn in (pre-AI) Photoshop from scratch, for interface purpose. The first is an actual application – mind the focus and guidance of attention and varying detail, the second one is just a draft, still lacking any such focus. This second one is quite similar to an AI generated image, as it lacks any purpose, thus, any reason to be there at all. (Besides that bit of light that is already in that image, it is offending in its neutrality. And there is really no point in publishing it, outside of this context.) [1] https://www.masswerk.at/eliza/ https://www.masswerk.at/eliza/ [2] https://www.masswerk.at/tennis-for-two/304a.html https://www.masswerk.at/tennis-for-two/304a.html