Free courses / Making Things With AI / Lesson 2
Describing what you want
Subject, setting, style, framing — and what to leave out.
A picture prompt works like the one you learned for text, with one extra part. Say the subject, the setting, the style and the framing, and then say what you do not want in it.
Compare “a dog” with “an old black labrador asleep on a worn sofa, afternoon light through a window, photographed from across the room, no people”. The second gets you something usable because you made four decisions instead of none.
The last part matters more than people expect. These tools add things to fill space, so saying what should not be there is often the edit that fixes a picture.
- Subject, setting, style, framing.
- Then say what you do not want in it.
- Change one thing at a time so you learn what did it.
PIXEL / THE COACH
I asked for a quiet beach and got forty-one people on it. I had not said no people. The beach was, to be fair, lovely.
Try it
Take a picture you generated in the last lesson. Change exactly one word of the description and run it again. Keep going until you can predict what each change will do.
Make a picture of a cup of tea going cold on a cluttered desk, late evening, lit only by a lamp, photographed from above. No people and no text anywhere in the picture.
Then paste it into whichever assistant you use — ChatGPT, Claude, Gemini or Copilot. Not sure which? Meet them all.
Progress is saved on this device.
Questions about “Describing what you want”Asked by readers, answered by us. Open to read them.
Open to load them.
Ask about this page and we will answer it here, where the next person can read it too.
Nobody else can answer you, and you cannot answer anybody else. This is not a forum — we would rather say that than run a room we cannot watch at three in the morning.
Your question goes up with the answer and without your name. Nothing identifies you, ever.
If we cannot answer it we will tell you why, and that one stays between us.