> have nothing “intentionally” placed (apart from tuning/tweaking the balance of the seeding algorithms).
I think you underestimate the intentionality that goes into developing procedural generation. Something like Dwarf Fortress isn't "place objects randomly" - it is layers upon layers of carefully crafted systems that build upon each other to produce specific patterns of outcome
By calling it out in my comment, I was trying to not underestimate it.
I guess what I'm saying is: Couldn't a world model with targeted training and thoughtfully tuned system prompts be directionally similar to the layered systems to produce specific patterns of outcome?