Raw LLM Responses

Claiming to have an eye for anatomy and proportions is wild, when looking at all the things he does have there is something very off about many of his proportions that is consistently inconsistent in his work. It sits in the uncanny valley or trying to be corrected but the exaggerations aren't really exaggerated enough to sell a sense of fluidity or presence or balance. Every single piece is missing at least one of three lines in the line of motion, line of balance, or line of purpose/context. When I have a client that commissions me for a piece, the man that showed his book and talked about prompting, That's a process that's is generally called creating thjings called thumbnails and in a good client/artist relationship, the artist will often have many ways on how they also interpret the 'vibes' of what they believe their client wants from them and what is being payed for. Personal rule of thumb is that I won't even present a thumbnail that I don't find satisfactory, but these thumbnails really are as basic do-it-in-less-than-a-minute, very abstract doodles that try to visualize and establish a context. They've very much like storyboarding images. When a client likes a few of them, then I'll refine them to look a bit more cohesive and less abstract, and then the client will typically pick their best while requesting certain changes (Fun fact, most of my clients actually have to be encouraged to make these).. once I have their favorite of the bunch, then I work on the actual final composition while keeping them informed all along any step of the way while making sure to leave myself ways to adjust certain things later if requested without having to rework a massive amount of the image. Looking at the picture you show him fussing and fawning and patting himself on the back over. Something AI struggles with actually showing but is good at faking is the last of the above. Any section of an image that would be in 'motion' should serve to provide context of the character, the environment, or the direction of wind.... and with the weird amorphous hair off to her right but the cape kind-of out back behind her... there's no clear line of context... where is the hair coming from? What is some of it even anchored to? Fabric doesn't have to be realistic in a stylized image but it should still serve the purpose of context. No real artist that has reached that level of actual skill on composition would ever make such careless mistakes in a piece they'd put so much time into, but AI doesn't have ANY intent behind resolving the quirks that come up in the creative process, it just scraps it entirely or warps things to try and shove them in the square hole provided by the prompt... If the only interesting aspect of looking at an image is noticing all the uncanny ways it's warped because a machine doesn't understand what fabric actually is and that it lays very differently or can be exaggerated in a dynamic way, what's the point of looking at it? I really miss being able to randomly browse places like Deviantart and stare at jaw-dropping compositions and seeing more and more little things the longer I looked as expressions and intentions and displays from the artist. It's what I strive for when I make my own art, even if most people never see or notice certain details, those that do look for them are rewarded and it's a way for me as the artist to communicate something human-to-human. Even edited AI images don't have that because they were GENERATED, not crafted.