1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

Session 10 of 24 — AI Builder

AI Builder · Session 10 of 24⏱ 30 min · 👨‍👩‍👧 Parent aware

📝 Chain Prompting

Build complex outputs by chaining multiple prompts.

🎯 Today's goal: Your child learns that the best AI outputs come from a sequence of connected prompts — not a single perfect instruction.

🔥 Warm Up — 3 minutes

Warm Up

"The most powerful way to use AI is not one big prompt — it is a chain of smaller prompts that build on each other. Each output becomes the input for the next."

"Today we build something complex by chaining five prompts together."

🤖 The Activity — 20 minutes

Main Activity

Today's project: create a complete guide to a topic your child knows well — using five chained prompts, each building on the last.

Prompt 1 — The skeleton:

📋 Type this exactly
Give me a structure for a complete beginner's guide to [child's chosen topic]. Just the headings — no content yet. 6-8 headings. Make the structure logical and progressive.

Prompt 2 — Fill the first section:

📋 Type this exactly
Using this structure: [paste the headings]. Write the first section only. Make it genuinely useful for a complete beginner. About 100 words.

Prompt 3 — Critique and improve:

📋 Type this exactly
Here is the first section of a guide: [paste it]. What is weak about this? What is missing? Rewrite it — better.

Prompt 4 — Add your personal knowledge:

📋 Type this exactly
Here is the improved first section: [paste]. I am an expert on this topic. Here is one thing I know that is not in this section: [child adds their own knowledge]. Incorporate this naturally.

Prompt 5 — Polish:

📋 Type this exactly
Here is the complete first section: [paste]. Make it flow better. Fix any awkward sentences. Keep it under 100 words but make every word count.

The final output is genuinely better than any single prompt could produce. Discuss: what made chaining better than one big prompt?

🧩 The Twist — 5 minutes

Deep Dive
📋 Type this exactly
What is the worst way to use chain prompting? Give me an example of a chain that would produce a terrible final output because of how it was structured.

Ask: What makes a good chain versus a bad chain? Answer: each prompt should refine or build on the previous output. A chain that drifts or contradicts itself produces garbage.

🛡️ Safety Moment
“Don't use AI to do your homework”

Chain prompting is one of the most powerful AI techniques. Used to build understanding — fantastic. Used to write your school essay section by section — still cheating. The technique is neutral; the application is what matters.

👨‍👩‍👧 Parent: read this aloud. Ask your child to repeat it back.
📝

Chain Builder Badge 📝

Built a complex output using five connected prompts — each improving on the last

👨‍👩‍👧

Parent Notes — tap to expand

▼
What they learned

Advanced prompting technique — chain prompting. The insight that AI outputs improve through iteration is one of the most practically useful things in the programme.

Questions to ask
  • What made the fifth version better than the first?
  • Where in real life would chain prompting be most useful?
  • Could you use this technique for a school project? How?
What to watch for

Children who master chain prompting often become significantly more capable AI users. The technique applies to almost every complex task.

Safety in context

The chained homework essay concern is real. Have the conversation explicitly: the technique is powerful, and using it to do school work is still doing school work for you.

← Session 9Session 11 →