World 6: Build the Future · Lesson 6.2.1

Your AI Invention

Everything a teacher needs to deliver this lesson — pick your grade's script below once you've read the background.

Learning Objectives

This is the capstone of World 6 — the lesson where a student stops learning about AI and instead uses everything AIQ has taught them to design something of their own. It introduces no new facts; its job is synthesis. Across the four age bands, by the end of the lesson a student should be able to:

Pick a real problem before thinking about the technology. Every version of this lesson opens the same way, at different vocabulary levels: a good invention starts with a problem someone actually has, not with "AI is cool, what can I use it for?"
Describe their invention in one clear sentence — who has the problem, what the problem is, and why it matters to them personally. Younger students can do this out loud; older students should be able to write it down.
Name at least one way their invention could go wrong or be unfair, and one thing they could do about it. For 5–10 year-olds this is as simple as "would this be fair to everyone?" — for 11–18 year-olds it should sound like an actual risk (bias in the data, who gets left out, what happens when it makes a mistake).
Explain their invention to someone else using a concrete example of a person it would help, not just a list of features — "Meet Sarah. She has problem X. My AI does Y. Now Sarah can Z" is the app's own template and works at every age.
(Ages 11+) Match an AI technique to the problem — recognizing, at least at a basic level, whether their idea needs something that classifies, predicts, generates, or recommends, and why that particular technique fits.

Because this is a capstone rather than a content lesson, success looks different from a normal AIQ lesson: there is no single correct invention, and grading (if you grade it at all) should reward the quality of the thinking a student shows at each stage, not the ambition or cleverness of the final idea. A simple, well-reasoned project from a struggling student is a bigger win here than an impressive-sounding one nobody can defend under a follow-up question.

Teacher Background

You do not need any AI or programming background to run this lesson, and — importantly — no student needs to actually build a working piece of software to complete it. "Invention" here means a designed idea: a problem statement, a rough plan for how AI would help, a look at who it might affect, and a way to present it (a drawing, a poster, a short pitch). Nothing in the app requires code to be written or a real model to be trained in this lesson; that already happened, at an appropriate level, in earlier World 6 lessons like 6.1.2 ("Train Your Own Model"). This one is about putting the pieces together into an idea a student can call their own.

The app walks students through the same three stages for every age group, just at very different depths:

1. Choose Your Problem. Students pick a real problem from one of four areas — school (bullying, tutoring, scheduling), the world (climate, endangered species, disaster response), home (recycling, energy, family scheduling), or community (accessibility, translation, elder care, traffic). The point of offering four categories is to make "pick a problem" concrete rather than paralyzing — a blank "invent anything" prompt is much harder for most students than choosing from a short, relatable list.

2. Design Your Solution. This stage has four parts: writing the problem down as one sentence (who, what, why); thinking about what data the AI would need and where it would come from; matching an AI technique to the problem (classification, prediction, generation, or recommendation — introduced for older bands only); and an ethics pass — could this be biased, who might it affect, what happens when it's wrong.

The single idea worth protecting if you protect nothing else: ethics is not a last step. The app deliberately puts "The Ethics" as one of the four equal parts of designing the solution, not as a review that happens after the fun part is done. If a class is short on time, resist the temptation to skip straight from "the AI part" to presenting — even one sentence of "who could this hurt, and how would you find out?" for every project is worth protecting.

3. Present Your Invention. Four components: a drawing or diagram of how it works, a short story naming a specific person it would help, a number that shows scale of impact (how many people, how much time saved), and actually sharing it with an audience — family, classmates, or the whole class. The app's own summary line for this lesson makes the priority explicit: caring about the problem and the people involved matters more than technical polish or fancy language.

Because this is a capstone, expect a wide range of ambition and technical depth even within one class — that's fine, and it's actually the intended outcome. A seven-year-old's invention might be "an AI that tells you if it's going to rain so you remember your umbrella," fully valid at that level. A seventeen-year-old's invention should look like a real (if unbuilt) technical proposal, with a named technique, a data source, and an honest limitations section. Grade for depth of thinking relative to the age band, not against a single fixed bar.

The lesson's practice activity (a quick sorting exercise before the quiz) reinforces one more idea worth knowing about: it separates the parts of building an invention that are "the AI part" — technique selection, model choice, MVP prototyping — from the parts that are entirely human judgment — defining the problem, planning data ethically, presenting, and measuring impact. This maps onto the course's recurring rule from earlier lessons: you direct, AI executes, you judge. Even in a lesson about "building with AI," most of the actual work students do is human thinking, not machine work.

Materials & Prep

A device with a browser and access to aiq.ph is the only strict requirement — the guided scenes and quiz run entirely in the app.

For the presentation step, plan for extra time and simple supplies: paper and markers or crayons (for a poster or diagram), or a shared slide/document if the class is 1:1 device. This is the one part of the lesson that benefits from being taken outside the app rather than left as a screen activity.
For Explorer and Builder classes: nothing beyond paper and something to draw with. No prior project or homework is assumed.
For Hacker and Architect classes: if you want to stretch this into a fuller project (see each age page's extension activity), index cards or sticky notes help with the ethics/stakeholder brainstorming step, and a shared document is useful if students are drafting a written problem statement or technical plan.
No accounts, sign-ups, or student data collection are needed to deliver the core lesson.

Because a few age bands' extension activities involve student surveys or brief outside research (see the Hacker and Architect pages), it's worth reminding students that any "research" for this project stays informal and classroom-scale — a show of hands, a quick chat with classmates or family — rather than anything that would need real consent processes or data handling of its own.

Common Misconceptions

"This capstone means students need to actually code and train a working AI model."

No building or coding is required to complete this lesson. "Invention" here means a designed idea — a problem statement, a plan for data and technique, an ethics check, and a presentation (poster, story, or pitch). Real model-building, at an age-appropriate level, already happened earlier in World 6; this lesson is about applying that understanding to a project of the student's own choosing, not repeating it.

"The best AI inventions come from the coolest or most advanced technology."

The app is explicit about this: the best inventions start with a real problem the student personally cares about, not with impressive-sounding technology looking for a use. A simple idea that clearly helps a specific person beats an ambitious one nobody can explain the need for.

"Ethics and fairness are things you check at the end, once the idea is mostly finished."

The app treats "The Ethics" as one of four equal design steps, alongside the problem statement, the data, and the AI technique — not as a final review. Asking "who could this affect, and could it be unfair?" while the idea is still being shaped catches problems that are much harder to fix once a student is emotionally attached to a finished plan.

"A good presentation means using technical vocabulary and sounding impressive."

The app's own summary for this lesson says the opposite: caring about the problem and the people it affects matters more than big words or perfect execution. The presentation template it teaches — name a specific person, describe their problem, explain what the AI does for them, and show what changes — works precisely because it's concrete and human, not jargon-heavy.

Pick your grade's script

🌈 Explorer — Ages 5–7 Full delivery script for this age group. 🔧 Builder — Ages 8–10 Full delivery script for this age group. 💻 Hacker — Ages 11–14 Full delivery script for this age group. ⚡ Architect — Ages 15–18 Full delivery script for this age group.

← All lesson plans ← AI + Creativity Mashup AI Career Paths →