Tell AI exactly what worked, what did not, and what to change — then ask it to revise. The more specific your feedback, the better the next version. Vague feedback like “make it better” produces vague improvements. Specific feedback like “shorten the intro and add a teaching analogy” produces exactly what you need.
The Red Pen Approach
Think of giving AI feedback the same way you would mark up a student’s essay. You would not write “needs work” in the margin and hand it back. You would circle the weak paragraph, explain why it does not land, and suggest what to try instead. AI responds to the same kind of specificity. It cannot read your mind, but it can follow precise instructions.
Most educators ask AI for content, get a mediocre result, and either rewrite it themselves or start over with a new prompt. Both approaches waste time. The better move is to give feedback within the same conversation. AI remembers the entire thread, so each revision builds on what came before.
Five Types of Feedback That Work
First, point to what you liked. “The second paragraph is perfect — keep that tone for the rest.” This tells AI which parts to protect so it does not accidentally change something that was already good.
Second, name what missed. “The opening sounds too formal. It reads like a textbook instead of a conversation.” AI needs to know what the problem is before it can fix it.
Third, give a direction. “Rewrite the intro as if you are talking to a fifty-year-old coach who just finished a Zoom session with her clients.” This gives AI a concrete target instead of making it guess what you want.
Fourth, use examples. “Here is how I would say this: [paste your version]. Now rewrite the rest of the article to match this style.” AI learns faster from examples than from abstract descriptions.
Fifth, iterate. Do not expect perfection on the second try. Each round of feedback gets you closer. Most educators find that two to three rounds of feedback produce content that needs only light editing — far less time than writing from scratch.
What This Means for Educators
As a course creator, learning to give good feedback to AI is a skill that compounds over time. Better feedback means better first drafts, which means less editing, which means more content published with less effort. It is exactly like training a teaching assistant — the investment in clear communication pays off on every single task.
The Simple Rule
After every AI output, say one thing you liked, one thing that missed, and one specific change to make. Then ask for a revision. This three-part feedback habit takes thirty seconds and transforms your AI from a mediocre first draft machine into a reliable content partner.