AI Basics
AI Basics
What AI is, the three kinds you already use, and a hands-on first exercise with a chat assistant.
Tutorial overview
What you will learn
- Explain what AI is in one plain sentence
- Identify the three types of AI you already use
- Complete a first productive task with a chat assistant and verify the result
By the end, you will have
- A working mental map of everyday AI and one completed hands-on exercise
Introduction
Artificial intelligence (AI) is software that handles tasks which normally need human judgment, such as understanding language, recognizing images, or ranking options. It is not a robot brain and it does not think the way people do. It finds patterns in large amounts of data and uses those patterns to make useful guesses. You already rely on AI every day: your email spam filter, your phone's face unlock, and the route your map app picks are all AI at work.
This tutorial gives you a practical map of the AI you already encounter, then walks you through your first deliberate, productive use of a chat assistant.
What you will build or practice
By the end of this lesson you will have three things in your notes:
- A list of AI systems you already use without thinking about them.
- One completed real task done with a chat assistant.
- A short personal checklist for verifying AI output before you trust it.
Before you begin
You need a web browser, a notes app, and a free account with any chat assistant such as ChatGPT, Claude, or Gemini. No coding, no math, no downloads. Plan for about thirty minutes.
Key concept
AI is not one technology. As an everyday user, you meet three broad families:
- Assistants and generative tools. Chat assistants, image generators, and writing aids that create new text, images, or code from your instructions.
- Recommendation systems. The software that decides which videos, songs, products, or posts you see next, based on what you and similar people chose before.
- Recognition and automation. Spam filters, speech-to-text, photo search, face unlock, and fraud alerts that classify things quietly in the background.
All three learn patterns from data instead of following hand-written rules for every case. That makes them powerful and flexible, but it also means they can be confidently wrong, so verification is part of using them well.
Step 1: Spot the AI you already use
Open your notes app and list every AI-powered feature you touched in the last 24 hours. Common answers: email spam filtering, autocomplete on your phone keyboard, map navigation, streaming recommendations, photo search by keyword, voice assistants. Label each one as an assistant, a recommender, or recognition. This exercise turns AI from a buzzword into a set of concrete tools.
Step 2: Run a real task through a chat assistant
Pick a small task from your actual life and hand it to a chat assistant with enough context to succeed. For example:
I am writing a short email to my landlord asking to fix a leaking
kitchen tap. Keep it polite, firm, and under 120 words. Mention that
I first reported it two weeks ago.
Notice the pattern: a clear goal, relevant details, and constraints on length and tone.
Step 3: Improve the result with a follow-up
The first answer is a draft, not a final product. Reply with a specific request: make it shorter, change the tone, add a deadline, or offer two versions. Iterating like this is the core skill behind prompt engineering basics, and it works with every assistant.
Step 4: Verify before you use anything
Read the output as a skeptical editor. Are the facts real? Does the tone fit? Would you sign your name to it? For anything factual, check at least one independent source. The habit is covered in depth in how to verify AI answers.
Practice exercise
Choose one recurring task in your week, such as summarizing an article, planning meals, or drafting a message. Run it through an assistant using the goal-details-constraints pattern from Step 2, iterate once, and note what the AI did well and where you had to correct it.
Common mistakes
- Treating AI output as finished work instead of a first draft.
- Giving one-line prompts with no context, then blaming the tool.
- Assuming the assistant knows current facts about you or recent events.
- Trying every new AI tool at once instead of getting good with one.
Check your understanding
- What are the three families of AI most people use daily?
- Why can an AI system be fluent and still be wrong?
- What three elements make a task prompt effective?
Key takeaways
- AI is pattern-finding software, not a thinking machine.
- You already use AI daily through assistants, recommenders, and recognition systems.
- Clear context plus iteration plus verification is the reliable workflow.
Next steps
Go deeper on definitions with What Is Artificial Intelligence?, then follow the AI beginner roadmap for a structured path.
Related resources
Newsletter or next lesson
Join the newsletter for one clear lesson each week, or continue to What Is Artificial Intelligence?