>>28617
Anatomical cross-sections are a real challenge, and I don't think AI will grasp them immediately—though major progress has been made with models like Wan, Grok, and Minimax. So, for now, I think we still need to use Photoshop to composite a photographic cross-section over the gap the AI ​​leaves in decapitation shots. Even when providing maximum anatomical detail, it’s still hard to get the AI ​​to understand the specific angle or orientation for that cut surface...
I don't get hung up on these details because they aren't deal-breakers for my scenarios—even though, I agree, it would be great if they were handled better :)