Write the brief

arqiv✳

Lesson · Design thinking & clear communication

Write the brief

Students learn that a good prompt is really a design brief, the same kind of document engineers and product managers write every day. A vague request gets a generic result; clear thinking gets a useful one.

Level 3 AI as collaboratorGrades 6–1245 minutesEngineering · Art · Any subjectPaper + 1 teacher screen

The big idea

The clearer your thinking, the better any tool, or any teammate, can help you.

Students will be able to, without AI:

  • Define a problem, audience, constraints and success criteria
  • Write a brief someone else could act on
  • Judge outputs against their own criteria
  • Explain why clear specifications matter in real jobs

Lesson flow · the 5 P’s

Think first. Then the tool.

Most of the lesson happens on paper. AI appears for one short, structured phase, and the teacher can run it from a single projector screen. How the 5 P’s work →

1

Ponder

10 min○ Screens off · paper only

Teams pick a real design challenge from school life.

  1. Choose one: a poster that gets more students to the science fair; a better lunch-line system; a sign that reduces hallway crowding.
  2. Who is it for? What problem does it solve?
  3. What must it include? What must it not do?
  4. How will you know it works? Write two success criteria you can check just by looking at a draft (for example, “the date and room can be read from 10 feet away” or “a new student could follow it without help”).
Say to students“In real jobs, people who can explain exactly what they need are the ones who get great results, from people and from AI.”
2

Plan

10 min○ Screens off · paper only

Teams write a one-page brief on the worksheet.

  1. Goal · Audience · Must have · Must not · Constraints (size, budget, time) · Tone or style in words · How we’ll judge it.
  2. Trade briefs with another team: could they build it from your words alone? Revise.
3

Partner

10 min● AI on · teacher screen or lab

Every team needs two results for its own challenge: one made from a vague one-line request and one made from its brief.

  1. With devices (a lab, or time on the teacher’s screen): run both prompts for every team. Each team sketches or notes its two results.
  2. One screen and not much time? Run both prompts for one team as a demo. Then each team “plays the AI” for another team: one quick draft from the one-line request alone (2 min), then a second draft from the brief alone, with no questions allowed (4 min). Both drafts go back to the team that wrote the brief.
  3. Students note differences between the vague and the briefed result.
[Your challenge in one vague line, for example: “Make a poster about the science fair.” / “Fix our lunch line.” / “Make a sign about hallway crowding.”]
[Paste the full team brief] Using this brief, design the [poster / lunch-line system / sign]. Explain how each part meets the success criteria.
4

Probe

7 min○ Screens off · paper only

Screens off. Each team scores its own two results against its own success criteria. (If another team played the AI, “the AI” below means that team.)

  1. Score each result 1–4 on each criterion.
  2. What did the vague prompt force the AI to guess?
  3. What did the AI still get wrong, even with the brief?
  4. What would you change in your brief next time?
5

Prove

8 min + homework○ Screens off · paper only

Teams present the journey, then reflect.

  1. Two or three teams present, 2 minutes each: our brief → what came back → what we changed → our final idea.
  2. Every team writes its reflection on the worksheet: where else is a clear brief useful? (Group projects, jobs, asking a teacher for help.)
  3. For homework or next class: build or sketch the final version by hand.

Printable

Student worksheet

Print one per team, or one per student if each student keeps a copy of the team brief. Everything students hand in is handwritten, so their thinking is visible.

Design brief

Clear thinking in, useful results out.

Names:Date:Class:
Goal: what problem are we solving?
Audience: who is it for?
Must have
Must not
Constraints (size, budget, time, materials)
Style or tone, in words
How we’ll judge success (two criteria we can check on a draft)
Scores (1–4 for each of our two criteria)
Vague result: ____ and ____ · Briefed result: ____ and ____
What the vague request made the AI guess, and what we’d change in our brief
Reflection: where else is a clear brief useful?

Assessment

Grade the thinking, not the output

Every row scores something students did themselves. An AI can produce a product; it can’t produce a student’s reasons.

Criteria4 · Excellent3 · Solid2 · Developing1 · Beginning
Problem definitionSpecific goal, audience and success criteriaClear goal and audienceVague goalMissing
Brief qualityAnother team could act on it without questionsMostly actionableSeveral gapsNot usable
EvaluationScores outputs fairly against criteria and explainsScores outputsOpinion onlyNone
Iteration & reflectionImproves the brief and connects to real workImproves the briefMinor changesNo reflection

Why this is hard to cheat

Students are graded on the brief, the scoring and the reflection, all done by hand. The AI output is something to score, not something a student hands in as their own.

Adjust it

Younger students: Use a fill-in-the-blank brief with pictures. The teacher compares a vague and a detailed drawing request.

Older students: Write a real product requirements document for a school app idea, including user stories and what the first version will leave out.

No devices at all? Teams play the AI for each other, as in Partner, so every team still gets two results to score. To show real AI output too, print a vague and a briefed result for each of the three challenges in advance.

Discussion questions

  • Why does a vague request get a generic result, from a person or an AI?
  • Who writes briefs in real jobs? (Engineers, product managers, architects, film directors.)
  • What’s the difference between telling AI what to do and thinking for it?
  • How could a clear brief help in a group project?

Extensions & connections

  • Careers: product manager, engineer, architect, UX designer.
  • See how engineers set goals for AI-designed parts on AI in engineering & robots.
  • Use the brief template for any project in any class.
  • Try the prompt builder on Talking to AI: turn the parts of a prompt on and off and watch the guessing disappear.

Put curiosity to work

Read it. Try it. Question it.

Explore new AI research, or take a paper-first investigation into your classroom.