I am still playing with this idea of a graphic novel. Consulting with an artist who has interest in my project, I was instructed the first thing I needed to do was create a script.
A script????!!?
Not knowing what I had gotten myself into, I referred to the online Gods to tell me what to do. And lo, a 'tube video was found that had almost play-by-play instructions on how to use ChatGPT, the @openai tool, to create a script. I told it to use elements of "the writer's journey" and the "Save the Cat" beats storytelling structure, fed it some character details, and off it went. Ok I admit it was not quite that easy - because it did go off, off the rails that is. The tendency for the AI was to come to resolution quickly or to state the obvious, without nuance. Granted I was using the free platform, 3.5, and maybe the enterprise or other more sophisticated iterations would not be so limited. But all I wanted was a quick script to check the box for my artists to consider.
A week and half later, I finally managed to get the script I wanted, with a lot of heavy editing and coaxing and reworking on my part. Let's call this first entree into the land of AI generation a bit of a learning curve. And definitely not as easy as "press button make go."
I shared the script and the artists agree it conveyed my vision, enough for them to begin work. There's the small matter of expenses. Here's where @kickstarter comes in. Ok, design a Kickstarter page. Check, did that.
When I asked for their review of the campaign, I was once again given consul - where's the mock up?
Ironic, I thought, that the artist is asking the author, who is trying to hire an artist, for the artistic mock up.
Ok, back to our AI frenemies..... After using four different sites, giving basic phrases and ideas, I got enough images to be able to use PowerPoint and the Design Maker feature to come up with this mock up. What do you think? Not terrible....not perfect either.... conveys enough of an idea for one page. I would not want to do that again for the other 120 pages.
I found the whole process fascinating. For the most part the AI did a decent rendition of the graphics. I was not able to get it to do exactly what I wanted, so the effort was a bit frustrating. I felt it was a good accelerator. Maybe it could produce everything exactly as I wanted if I could communicate with it more clearly in the manner it requires.
Question is: was my experience the result of my own limitations or the limitations of AI to be completely creative?
コメント