👁️
SEE
Recognise objects, faces, handwriting and animals inside images.
PROJECT 3 · UNDERSTANDING ARTIFICIAL INTELLIGENCE
ARTIFICIAL
ARTIFICIAL
INTELLIGENCE
Machines that can learn, notice patterns, and help people solve problems.
AI is becoming part of how we learn, communicate, work and create. But what exactly is it, and how does it actually work?
WHAT AI DOES ALL DAY
- LEARN
- CREATE
- THINK
- ASK
- DISCOVER
- SOLVE
Every section of this page answers one question. By the end you should be able to explain AI to your parents in one minute.
14
sections
5
questions to answer
0
lines of JavaScript
SECTION 01
What Is Artificial Intelligence?
Artificial Intelligence, or AI, is technology that allows computers to perform tasks that normally require human intelligence — understanding language, recognising pictures, answering questions, translating, predicting and creating.
A computer program normally follows rules a person wrote. An AI system is different: it is shown many examples and it works out the patterns for itself.
🎤
LISTEN
Turn speech into text and understand spoken commands.
💬
COMMUNICATE
Answer questions and hold a conversation.
💡
LEARN
Find patterns in very large amounts of information.
✏️
CREATE
Help produce text, images, music, video and code.
SECTION 02
You Probably Used AI Today
Most people think AI means robots. In reality it is already sitting quietly inside the apps you opened this morning.
face unlock. All of it is AI.
📱
Smartphones
Face unlock, voice assistants, camera modes and autocorrect.
▶️
YouTube
Recommends the next video from what you watch.
🎵
Music apps
Suggest songs that match what you already like.
🗺️
Maps
Calculate routes and predict traffic.
✉️
Detects spam and suggests the next word.
📷
Social media
Ranks posts, detects faces, suggests content.
🤖
AI assistants
Explain concepts, brainstorm ideas, help you study.
🔤
Translation
Move a sentence between languages in a second.
How many times have you used AI today without knowing it?
SECTION 03
How Did We Get Here?
AI did not appear in 2023. It is the result of seventy years of slow work, and of three things arriving at once: faster computers, more data, and better methods.
1950s
The big question
Scientists began asking whether a machine could think, and whether a computer could imitate human intelligence at all.
1960s–
1980s
Early AI
Researchers built programs that followed rules and solved narrow problems. Computers were slow, expensive and could not handle much information.
1990s
Better computers
Machines became far faster. AI improved at recognising patterns, processing information and playing games.
2000s
More data
The internet produced enormous amounts of digital information — and data is exactly what AI systems learn from.
2010s
Machine learning grows
Image recognition, speech, translation and recommendations improved sharply. Machine learning became the centre of modern AI.
2020s
Generative AI
AI could generate text, images, code, audio and video, and people could talk to it in ordinary language.
AI moved from something a few researchers used to something millions
of ordinary people use.
SECTION 04
How Are AI Models Trained?
Nobody writes a rule for what a mango looks like. Show a child enough mangoes and they start noticing the shape, the colour, the size, the skin. Later they recognise a mango they have never seen before.
AI training works in a similar way. Instead of teaching every rule by hand, developers provide many examples and let the system adjust itself until its answers get better.
THE TRAINING JOURNEY
DATAexamples to learn from
TRAININGprocess the examples
PATTERNSwhat repeats
MODELthe trained system
TESTINGsomething it has not seen
AI OUTPUTa prediction
01
Collect data
Text, images, audio, numbers, video — many examples of whatever the model must recognise.
02
Training
The computer works through the examples looking for what repeats: shapes, ears, faces, colours, textures.
03
The model
The trained result. A model holds the patterns and uses them to make predictions.
04
Testing
Show it something new. Is this a cat or a dog? Developers measure how often it is right.
05
Improvement
Better data, more training, adjusted methods, test again. This loop repeats many times.
SECTION 05 · A WORKED EXAMPLE
Let's Train a Fruit AI
We want a model that can tell a banana from a mango from an orange. We collect the training data, we train, then we show the model a photo it has never seen.
- 1,000 banana images
- 1,000 mango images
- 1,000 orange images
The model does not "know" the fruit the way you do. It compares patterns and gives its best prediction.
NEW IMAGE → PREDICTION
MANGO
Confidence: 94%
94% confident is not the same as 100% correct.
That gap is the whole reason we check
AI answers.
SECTION 06
How Can AI Understand Our Questions?
Assistants like ChatGPT are large language models. They were trained on enormous collections of text, and what they learned is how language tends to continue.
Cameroon is a country in ___
The model predicts which words are likely to come next. Repeat that at enormous scale and it can explain, summarise, translate and write.
- Answer questions
- Summarise
- Explain topics
- Write
- Translate
- Generate ideas
- Help with code
AI predicts patterns. That does not automatically make every
answer correct.
SECTION 07
What Can AI Do Today?
Not one day in the future — today, in these eight fields.
📚
Education
Explain lessons, generate practice questions, summarise topics, support teachers.
🏥
Healthcare
Help professionals analyse information and spot patterns in medical data.
🌱
Agriculture
Analyse crops, weather information and farming conditions.
🚗
Transportation
Navigation, traffic prediction and some automated vehicle systems.
💳
Finance
Detect suspicious transactions and analyse financial information.
🎨
Creativity
Assist with images, music, writing, video and design.
💻
Programming
Explain code, suggest code, find errors, help developers learn.
🔤
Language
Translate between languages and support communication.
SECTION 08
Why Do We Use AI?
Seven reasons, and none of them is "because it is new".
SPEED
Process very large amounts of information quickly.
PATTERN RECOGNITION
Notice patterns a person would struggle to see in time.
AUTOMATION
Handle repetitive tasks without getting tired.
CREATIVITY
Generate options and help explore possibilities.
ACCESSIBILITY
Speech recognition, image descriptions and translation open access for people with disabilities.
PERSONALISATION
Adapt recommendations or lessons to one particular learner.
RESEARCH
Analyse large datasets and explore difficult problems.
A calculator did not replace mathematics.
The internet did not replace learning.
AI should not replace thinking.
The best way to think about AI is as a tool that increases what humans are already capable of doing.
HUMAN + AI
→
BETTER POSSIBILITIES
SECTION 09
How Can Students Actually Use AI?
Eight things you can try this week. Notice that not one of them is "give me the answer".
work yourself.
1
Explain a hard topic
Explain photosynthesis to me like I am 12 years old.
2
Use AI as a tutor
Teach me fractions step by step and give me one question after each explanation.
3
Make practice questions
Give me 10 questions about the Solar System. Do not show the answers until I finish.
4
Improve your writing
Check my paragraph for grammar mistakes and explain what I should improve.
5
Brainstorm projects
Give me five Scratch game ideas about protecting the environment.
6
Understand code
Why is my HTML heading not displaying correctly? Here is my code.
7
Prepare for exams
Create a revision plan for my mathematics exam in two weeks.
8
Learn something new
Teach me the basics of robotics, starting from what a sensor is.
SECTION 10
Better Questions = Better Answers
A prompt is the instruction you give an AI system. A weak prompt gets a vague answer, and it is not the AI's fault.
WEAK PROMPT
Tell me about planets.
BETTER PROMPT
Explain the eight planets to a 12-year-old. Give their order, distance from the Sun, and one fun fact about each.
The formula for a good prompt
WHAT I WANT
+
CONTEXT
+
HOW I WANT IT
WORKED EXAMPLE
Teach me HTML headings
+ I am a beginner
+ use simple examples and give me one exercise
SECTION 11
Use AI To Learn, Not To Copy
Two students used the same tool for the same homework. Only one of them learned anything.
STUDENT A
- Copies the AI answer
- Submits it
- Learns nothing
STUDENT B
- Asks for an explanation
- Studies the answer
- Checks the information
- Writes it in their own words
- Actually learns
BE STUDENT B.
AI can be wrong
AI sometimes produces information that sounds convincing but is incorrect, incomplete or out of date.
- STOP
- CHECK
- VERIFY
- Ask where the information came from
- Compare it with your textbook
- Check a trusted website, ask your teacher
- Ask yourself whether the answer makes sense
Never confuse confidence with correctness.
Think before you share
An AI tool is not a private notebook. Never type these into one:
- Passwords
- Home address
- Family details
- Banking details
- ID numbers
- School records
Is using AI cheating?
It depends on how you use it and on the rules your teacher gives you.
GOOD USE
Explain a concept · practice questions · feedback · brainstorming · learning a method
BAD USE
Copying an assignment · using it in a test · submitting work you do not understand
AI should support your learning, not replace it.
SECTION 12
Explain AI At Home
"AI is technology that learns patterns from information, then uses
those patterns to do things like answer questions, recognise
images, translate languages or make recommendations."
Then give them the picture: imagine a student who studies millions of examples instead of one textbook. That student becomes very good at recognising patterns and answering certain kinds of questions. AI works somewhat like that — although a computer does not think the way a person does.
Finish with this: AI can help us learn faster, but we still need teachers, books, practice and our own thinking.
SHOW THEM, DO NOT ONLY TELL THEM
Demo 1 · Learning
Explain electricity to a 10-year-old using a water analogy.
Demo 2 · Language
Translate this sentence into French and explain the grammar.
Demo 3 · Creativity
Give me a story idea about a young inventor from Cameroon.
Demo 4 · Education
Create five mathematics questions for a student learning percentages.
Demo 5 · Programming
Explain the difference between HTML and CSS to a beginner.
SECTION 13
What Might AI Look Like Tomorrow?
Nobody can promise the future. But these directions are already being built.
AI tutors
Personalised learning assistants that move at each student's own pace.
Smarter healthcare
Helping doctors and researchers understand diseases more quickly.
Smarter agriculture
Better predictions about weather, crops, pests and soil.
Robotics
AI-powered machines taking on more complex physical tasks.
Scientific discovery
Analysing data to find new materials, medicines and methods.
New careers
Some jobs will change and new ones will appear — possibly ones that do not exist today.
for here.
What could AI mean for Africa?
The tools were mostly built elsewhere. The problems they could solve here are ours to name.
We should not only consume AI. We can learn to build with it.
Agriculture
Helping farmers read crop and soil conditions.
Education
Making learning materials reachable in more places.
Healthcare
Helping professionals analyse health information.
African languages
Tools that understand and translate our own languages.
Entrepreneurship
Helping young people research, design and market a business.
Local problems
Solutions designed for African communities by people who live in them.
SECTION 14
Will AI Replace Humans?
AI will change many jobs and many tasks. But the more useful question for a student is: how do I prepare to work with it?
AI IS GOOD AT
- Processing large amounts of information
- Repeating tasks without tiring
- Detecting patterns
- Generating possibilities
- Working quickly
HUMANS ARE GOOD AT
- Understanding context
- Caring for other people
- Creating meaning
- Making ethical decisions
- Choosing which goals matter
- CREATIVITY
- CRITICAL THINKING
- COMMUNICATION
- PROBLEM SOLVING
- TECHNOLOGY
- TEAMWORK
- CURIOSITY
- LEADERSHIP
HUMAN INTELLIGENCE + ARTIFICIAL INTELLIGENCE
The future may depend on learning to use both well.
Try AI This Week
DAY 1
Use AI to explain something you found difficult at school.
DAY 2
Ask AI to quiz you on a subject you are revising.
DAY 3
Use AI to brainstorm a Scratch or HTML project.
DAY 4
Find one mistake in an AI answer by checking another source.
DAY 5
Teach a friend or parent one thing you learned about AI.
AI IS POWERFUL. KNOWING HOW TO USE IT IS MORE POWERFUL.
The future will not belong only to people who have access to AI. It will belong to people who know how to ask good questions, think critically, create responsibly, and use technology to solve problems that matter.
Start Learning AI