Up until now I had not really used any of these AI-based image generation tools. Since I'm dusting off my writing habit, what better time to start playing around with images created using generative AI (GenAI) then when I am making new posts?
Better then stock photography and I'm building my skills for learning how to get these models to do my bidding 😀
Here was the prompt flow for this one:
Create a sketch in black and white of a simple robot drawing in a sketch book.
Make the style as if it was drawn with a charcoal pencil, with a bit more of a cartoon feel.
Make the robot a bit more human and add the word "Paul" to the picture it is drawing in the sketchbook
Didn't quite get the "Paul" part right, but it tried.
When I asked it to describe the image, this is what is responded with:
The image is a rough charcoal sketch depicting a robot with human-like features. The robot is sitting at a table and is engaged in drawing or writing in an open sketchbook. Its design is somewhat cartoonish, with a head that has a semblance of a human face and articulated limbs, giving it a more humanoid appearance. The robot is holding a pencil and appears to be writing the word "Paul" in the sketchbook. The background of the sketch is minimal, emphasizing the robot and its artistic activity in a spontaneous and less polished style.
As you work with the various AI tools and models that work with natural language, you start to see where it doesn't quite understand what you are asking.
This makes learning how to create prompts a bit of an art form.