Part I: The Visual Creator | Chapter 3
Generating and Controlling AI Images

The Goal
By the end of this chapter you will be able to make a set of images that belong together, through a process you can repeat instead of by luck. You will control subject, setting, composition, light and style, spot the typical errors of image generators, and know when a picture must be a real photograph instead of an AI image. You will make the set in ChatGPT, finish it in Snapseed and add real text in Canva.
Understand: the six layers of an image prompt
An image prompt is not one sentence. It is a set of instructions, like a shot list in film. Every prompt has six layers you can control:
The more precise each layer, the more control you have. A vague prompt gives a generic picture.
Continue with Missing Linkk Learn
The rest of this chapter is for members. Early access is free until 31 March 2027: every chapter, every prompt, and your own results saved on your phone.