4
14
I'm writing this mainly to answer for the journey of these two months, the problems I ran into, and what comes after.
Today marks exactly two months since I started creating in this direction. Seventeen works — eight of them are things that happen in the main world, exactly matching eight weeks. So, yeah — I'd say that counts as not letting you down.
I've always been writing small, interesting stories, and I believe the current tone is still at the point I like — a certain blend of Japanese-style fantasy and classic fantasy. And one layer deeper, I've actually been circling one basic framework the whole time: the hero, the princess, and D (dragon / demon).
I've always wanted to make stories like "opening a magic potion shop in this world" — that kind of thing. But I don't think my ability is enough to make it genuinely interesting — or is it the limits of the LLM itself? But for now, guess I'll keep being a hero — beat the dragon, save the princess.
What I'm making isn't a world you have to do homework for. Can a world have stories? Yes, it can. But it isn't suited to being LLM-driven — the LLM just can't handle too many complex twists. And I don't think I can portray those political, diplomatic, and religious conflict elements that are already complicated in reality. I just want to stay within what I can do.
All our stories right now are inside the first week of the Rift. The timeline will move forward, but not now. If a follow-up story appears after a story, it basically takes the main route of the previous story, and moves forward on the premise that it reached a good ending. We will reach the world's next step — but not now.
Every choice we make leads to a different next step. I think this is the key to roleplay in LLM-driven stories right now. Maybe what we're doing right now is actually the most orthodox Slay the Princess — we choose, we get a response, then we choose again. That's just like the difference between games and traditional art: interactivity. And our interactivity takes another route.
Some people say AI has no feelings. Maybe. But the feelings that appear in our stories are real.
Same as I said — one key point in how I design my works is that you absolutely don't need to read any worldbuilding to use them normally. But I did make basic lorebooks, and two of the more important ones I rebuilt — just in case someone actually wants to read what the world is:
And one more, an appearance-side rebuild:
Again, you don't need to know any worldbuilding to use my works normally. These are just for your interest.
The above is a summary of two months. Below are some interesting things I've found.
So it seems like a lot of people like making different personas for different stories, but I've only ever had one: LOKssr, a normal human.
So you can see, my stories are actually often better suited to freer starting points and worlds. I've gotten some comments about a lack of drive, and I've been trying to find a balance between them. Maybe this is what I'll mainly need to think about going forward.
I'm still a newbie. But two months have passed, and any of my mistakes or wrongdoings can no longer use "I'm a newbie" as an excuse. So feel free to point out what you're unhappy with in my works. Thank you very much.
Created Oct 30
Updated Oct 30
Published Oct 30