← All use cases

Use case P40 · Human

Returning cartoonist uses ChatGPT to turn her own six-panel script into a comic in under two hours

GeekMum ·

The short version

What they did

GeekMum returned to making cartoons after more than ten years away. She tested a generic four-panel prompt, then planned her own six-panel comic, specifying the setting, positions, dialogue, effects and captions before using ChatGPT’s image generation. She completed a first draft without revisions, but disliked its line breaks and limited editing control; it drew the wrong kind of saxophone and missed some intended hobbies. This is her personal cartoon experiment, not a client project or a finished commercial comic.

Result (as claimed)

“Is it what I wanted? No. But it’s done.”

“But it took less than 2 hours from idea to image, so that’s a win in my book.”

Prompt

Their prompt:

create a four panel comic that you think I’d enjoy based on what you know about me

This was her earlier four-panel test. For the six-panel comic she supplied a script and detailed layout directions, which the extracted article does not reproduce as a complete prompt.

For a six-panel comic:

A fuller version to copy:

I’m returning to cartooning and want to turn my own script into a six-panel comic. My script, panel by panel, is [paste setting, character positions, dialogue, sound effects and captions]. Character descriptions are [details]. The visual style is [line weight, colors and borders].

1. Check that each panel has enough information and ask about any missing detail before drawing.
2. Create the whole strip as one image, following the supplied panel order. Keep character appearances and important props consistent.
3. Use my dialogue and captions exactly, with readable speech bubbles and line breaks.

Keep the tone of my script. Do not add jokes, dialogue or hobbies I did not include. If the text or a prop is too difficult to render clearly, flag it rather than claiming the image matches perfectly. I’ll review the first draft before requesting changes.

Do it yourself

These steps are my suggested way in, not GeekMum’s setup.

What you need

  • A chat assistant that can make images, such as ChatGPT or Gemini.
  • Your own comic idea: a few characters, a setting and a punchline.
  • A free image editor for fixing text, such as Canva.

Start simple

  1. Write your script panel by panel: who is where, what they say, any sound effects or captions.
  2. Describe each character once in a short block: age, clothes, hair, one item they always have. Paste it into every request.
  3. Ask for the whole strip in one image. Keep each speech bubble under about 10 words.
  4. Ask for one change at a time: “Same image, but panel 3 shows an alto saxophone.”
  5. Save each version you like before you ask for the next change.

Take it further

  • Make each panel as its own image and lay them out yourself, so one bad panel doesn’t sink the strip.
  • Ask for the art with empty speech bubbles, then add the lettering in an editor. You control the line breaks.
  • Keep a style note (line weight, colors, borders) and reuse it for a series.

Check your work

  • Read every bubble word by word. Image tools still misspell and drop text.
  • Check props and details against your script, panel by panel.
  • Don’t upload photos of real people or other artists’ characters without permission.

Source checked. The link, author, date and quote were confirmed against the source. The result itself has not been reproduced. Their prompt is quoted from the source; any fuller version is written for this library. Business size: Individual. Results are quoted as the source claimed them and never improved. Spot an error or a broken link? Email charlie@cbarmorecpa.com.

More in Crafts, art and photos

New cases go out every Saturday in Inside AI.