AIForge

Learn how AI really works by building it: train a classifier, catch bias in the data, and become a sharp digital citizen who can spot a deepfake.

Level 1

For grown-ups — progress report →A private, on-device snapshot mapped to the skill areas practiced. Export a CSV or print — nothing leaves this device.

Train-a-Classifier

Show an AI some labelled examples and watch it discover a rule — no coding. See the rule it learned and why it decides, then find where its training data was unfair and fix it.

Train an AI →

Concepts — choose a kit

16 kits from “what is AI” to responsible-AI synthesis — each a short round of questions with hints when you need them. Your progress stays on this device.

Kit 1: What Is Artificial Intelligence? What is AI · grades 4-5 · 25 questions Kit 2: How Machines Learn: Patterns, Data, and Models Machine learning · grades 4-5 · 25 questions Kit 3: AI Ethics & Bias: Fairness in Algorithms Ethics & bias · grades 5-6 · 25 questions Kit 4: Data Privacy & Digital Citizenship Privacy & digital citizenship · grades 5-6 · 25 questions Kit 5: Computer Vision & Image AI Computer vision · grades 6-7 · 25 questions Kit 6: Natural Language Processing & Language AI Language AI · grades 6-7 · 25 questions Kit 7: AI and the Future of Work & Society AI & society · grades 7-8 · 25 questions Kit 8: Building Responsible AI Systems Responsible AI · grades 7-8 · 25 questions Kit 9: Machine Learning Basics Machine learning · grades 5-6 · 25 questions Kit 10: AI Ethics & Bias Ethics & bias · grades 6-7 · 25 questions Kit 11: Natural Language Processing Language AI · grades 6-7 · 25 questions Kit 12: Computer Vision Fundamentals Computer vision · grades 7-8 · 25 questions Kit 13: Cross-Topic Connections Review & synthesis · grades 5-6 · 25 questions Kit 14: Real-World Applications Real-world · grades 6-7 · 25 questions Kit 15: Misconceptions & Reasoning Reasoning · grades 6-7 · 25 questions Kit 16: Advanced Synthesis Review & synthesis · grades 7-8 · 25 questions

More AI challenges

Hands-on AI-literacy puzzles and extra practice modes:

Bias DetectorInspect a skewed dataset, name the bias, and fix it. Feature PickerPick the clue that sorts the examples most cleanly. Confidence MeterDecide: let the AI call it, or send it to a human? Train / Test SplitTest the AI on examples it has never seen. Mixed practiceRevisit questions from the kits you’ve played. AIForge SprintA calm speed round on what you’ve mastered. AI Quest mapA path through all 16 kits.

Play together

A quick AI-literacy face-off for 2–4 players — same device, or an online room by code.

AI Duel (pass-and-play) 2–4 players take turns on one device, or open an online room by code.

Same idea, another world

Something you practice here shows up somewhere else too — that's real learning that travels.

A skewed training set biases an AI the same way an unrepresentative sample skews a statistic.The shared idea: sampling bias — an unrepresentative sample skews the conclusion (a biased model or a biased statistic)Try DataForge →
AiForge mascot

Meet the cast — the characters who teach this

Each one stands for an idea you’ll practice. Tap a character to read their illustrated story and hear their audio drama.

Ensemble stories — the cast together

Explore all the stories →

📚 Books for AIForge

Read the story, print a workbook, or color the cast — free PDFs to read offline or print at home.

Browse all books →

Kids also liked

PlayNeuralQuestAppNeuralQuestAppAiForgePlayWonderForgePlayTableForgeAppModelQuest

Privacy & how it works