In the Hollywood context, what we do is called "fine-tuning" models. Let's say you wanted a specific dragon in a production. You could train a model to understand who that dragon is, how it moves, what it looks like - what it looks like really close up and really far away. And when you point the dragon this way and you tell it to fly upward toward the sunset, it'll understand what that means.
Every image generator is unique, but I've spent enough time with them to recognize some patterns. There are certain things AIs are prone to mess up -- fingers on hands, overlapping elements, extremely fine details, etcetera.