You've probably heard the term Artificial Intelligence - or AI - dozens of times this week. It's in the news, it's in adverts, and people at work won't stop talking about it.
But what actually is it? Let's clear up the mystery.
Before we define anything, here's a surprise: you're already an AI user. You have been for years.
AI isn't some futuristic robot. It's software that's already woven into your daily routine.
The average person interacts with AI-powered systems between 20 and 30 times per day - from unlocking their phone in the morning to getting a playlist recommendation at night.
Here's the simplest definition:
Artificial Intelligence is software that can learn from experience and make decisions, rather than just following fixed rules.
Let's make that concrete with an example.
A traditional calculator follows exact rules: "If the user presses 2 + 2, display 4." It will never learn, never improve, never do anything it wasn't told to do. It's brilliant at maths but can't do anything else.
A spam filter in your email doesn't have a fixed list of spam words. Instead, it learns by looking at millions of emails that humans have marked as spam or not spam. Over time, it spots new patterns that no programmer ever wrote rules for. It gets smarter with experience.
That ability to learn and improve is what separates AI from regular software.
Think about your email inbox. You probably haven't manually marked spam in ages, yet your spam folder catches new tricks from scammers every week. That's AI quietly learning in the background. Can you think of another example where software seems to get better on its own?
Not all AI is created equal. Scientists generally talk about two types:
Narrow AI is designed to do one specific task very well. Your phone's face recognition is narrow AI - it can identify faces, but it can't write a poem or drive a car.
Every AI you interact with today is narrow AI:
General AI would be a machine that can think, reason, and learn any task a human can - and possibly better. It would switch between writing music, diagnosing diseases, and having a philosophical debate, just like a person can.
This doesn't exist yet. Some researchers think it might take decades. Others think it may never happen. The AI you see in films like Ex Machina or Her is general AI - impressive but fictional for now.
What is the key difference between narrow AI and general AI?
AI is everywhere. Here are some examples you might not have thought about:
| Area | How AI Helps | |------|-------------| | Healthcare | Spotting early signs of cancer in medical scans | | Agriculture | Drones identifying which crops need water | | Banking | Detecting fraudulent transactions in milliseconds | | Retail | Predicting which products will sell out next week | | Transport | Optimising delivery routes for thousands of parcels | | Entertainment | Creating personalised playlists and recommendations |
The common thread? AI is brilliant at finding patterns in large amounts of data - far faster than any human could.
AI systems can now detect certain types of cancer from medical images with accuracy matching - and sometimes exceeding - that of experienced doctors. In one study, an AI identified breast cancer in mammograms 11.5% more accurately than human radiologists.
Which of these best describes what AI is?
AI is a tool - like a hammer or a calculator. It's incredibly powerful when used well, but it doesn't "think" the way humans do. Throughout this programme, we'll explore what AI can and can't do, so you can use it confidently and responsibly.
Which of these is an example of AI in everyday life?
Now that you know what AI is, the next lesson dives into how machines actually learn. We'll explore the fascinating process that turns raw data into intelligent behaviour.