AI Explorers · lesson 2 of 24 · 25 min
Before lunch today, you may have used AI five times without noticing. A map app picked your route. Your phone fixed a typo. A video app chose what to show next. Let’s find it all.
AI hides in plain sight
Most AI does not look like a robot. It lives quietly inside apps and devices. Here are common places you may meet it:
- Recommendations: video, music and shopping apps suggest what you might like next based on what you and people like you have watched or bought.
- Maps and directions: navigation apps predict traffic and pick a faster route.
- Typing help: autocorrect and word suggestions guess the word you meant.
- Photos: cameras find faces to focus on, and photo apps can group pictures by who is in them.
- Voice assistants: smart speakers turn your speech into text, figure out what you want, and answer.
- Safety: banks flag strange purchases, and email apps catch spam and scam messages.
Inputs, decisions, outputs
Every one of these systems follows the same simple pattern:
| Step | What it means | Example: music app |
|---|---|---|
| Input | Information the system takes in | Songs you played, skipped and liked |
| Decision | The AI’s guess or choice | “This listener likes slow R and B on weekend mornings” |
| Output | What you see or hear | A new song starts playing |
Once you can name the input, decision and output, you understand the shape of almost any AI system. This pattern will come back again and again in this course.
Helpful, and sometimes not
AI in daily life saves time. A route that avoids a traffic jam gets you to practice on time. A spam filter keeps scam texts away from your grandmother.
But it is fair to ask questions too. A recommendation app is trying to keep you watching, which is good for the company but not always good for your sleep. A map app needs to know where you are. Autocorrect sometimes changes your friend’s name into a random word because the name was not in its training data.
Good AI explorers do both: they enjoy what AI does well, and they notice who benefits, what data is being collected, and where it makes mistakes.
Here is a habit worth building. Whenever an app seems to read your mind, pause and ask: what did it learn from? A video app that shows you more basketball clips probably noticed how long you watched the last ones. A shopping app that suggests a phone case knows you just bought a phone. Once you start asking this question, AI stops feeling like magic and starts looking like what it is: patterns in data.
Try it
- Take out your “AI / Not AI” list from the last lesson.
- Pick three items from your AI column.
- For each one, draw a small three-box chart: Input → Decision → Output. Fill in each box in your own words.
- Under each chart, write one way it helps you and one question or worry you have about it.
- Bonus: ask a family member which AI tool they use most, and fill in a chart for theirs.
Check yourself
What are the three parts of the pattern most AI systems follow?
Input, decision, output. The system takes in information, makes a guess or choice, and shows a result.
In a spam filter, what is the input?
The email or text message itself, including its words, sender and links. The output is “spam” or “not spam.”
Why might autocorrect change an uncommon name?
The name probably was not common in the examples it learned from, so it guesses a more familiar word.
Key words
- Recommendation system: AI that suggests things you might like.
- Input: information a system takes in.
- Output: the result a system gives back.
- Spam filter: AI that sorts out junk or scam messages.
Previous lesson: What counts as AI?
Back to AI Explorers
Next lesson: What AI can’t do (yet)