gast@schmalzl.tirol — A good AI task needs a goal, context and examples
Style
?
PS gast@schmalzl ~C:\Users\gast/en/blog/2025-05-04-ki-aufgaben-klar-formulieren

A good AI task needs a goal, context and examples

By · Written: · Archive discovery date:

#KI #Prompts #Lernen

4 May 2025 · Archive date

"Make this better" can mean many things: shorter, more precise, friendlier, clearer or more detailed. If the goal is missing, the AI has to guess.

A good brief therefore explains the goal first. Then comes the information that is really needed for that goal. Finally, an example shows roughly what the result should look like.

The same task, described more precisely

Instead of "Write something about my plant shelf", an invented task could read: "Describe a shelf with three levels for private building instructions. Write for beginners, use short paragraphs and explain the structure first. Measurements are still missing: mark them as open details instead of inventing any."

Now purpose, audience, order and how to handle missing information are all named. A sample paragraph of your own shows the AI your writing style.

Headings such as Task, Background and Output make a longer brief easier to read. Whether you use Markdown or another format for this matters less than clearly separating the content.

Proofreading is part of the task

After the first answer, a second, narrower question is worthwhile: which statements are backed up? Which details are missing? Does the order suit the audience? A follow-up revision can then deal with these points directly.

A new AI answer is not an independent check of the truth. Figures, sources and practical details still have to be traceable. The plant shelf example works precisely because the open measurements stay visible. If information is missing, the AI should ask or mark the gap.

Original finds

These links were the starting point for this post. The assessment in the text is my own; the linked posts are not my own tests.

About the date: This post was written on 11 October 2026. The archive date is the date of the earliest related find in the collection, not of a publication at that time. Technical assessments and sources added later refer to the time of writing.