“Artificial intelligence” might sound like something from a movie, but actually, it’s all around us and becoming more common every day. If you’ve ever heard someone ask Siri what time it is or finished watching something on Netflix and had it suggest something for you to watch next, you’ve seen the work of artificial intelligence.
Artificial intelligence, or AI for short, is basically what it sounds like: It’s intelligence that’s human-made, something that can think like humans do without being a living creature like we are. AI is created using computers. If you’ve seen a movie, TV show, or video game with a robot character in it, you’ve seen an example of very advanced AI: It’s the technology that lets these robots act just like humans do. AI in real life isn’t quite that advanced yet, but it can help us to find, compare, and discover information in lots of different ways. First, though, we have to teach it lots of information, and we do that through a process called machine learning.
Machine learning is the process of teaching a computer (a machine) so that it learns to do something. The process usually starts by giving the computer lots of information, or data. This data might be labeled: For instance, you might give the computer lots of pictures of cats and dogs, with each one labeled as “cat” or “dog.” Then, you could start giving the computer pictures that aren’t labeled and see what it has learned: After looking at hundreds of pictures of cats and dogs, can it tell whether a new picture is a cat or a dog? It might not know right away, but when it’s wrong, the computer can learn from that and get smarter.
The goal of machine learning is to teach the computer to solve problems by connecting different ideas. That’s a lot like how human brains work, too. When you’re learning something new or trying to figure something out, you’ll often try to connect it to something you already know. For example, if you’re reading a book and you come across a big word that you don’t understand, you might try to figure it out by looking at the words around it and thinking about what words would usually fill in the blank between them.
Artificial intelligence is a whole area of study, while machine learning is only one part of how people make things with AI. To think of it another way, AI is like a subject in school (like “math”), while machine learning is one topic in that subject (like “multiplication”).
We come up with new ways to use AI every day! Virtual assistants like Siri and Alexa rely on AI. Video games use AI, too: It’s what tells the non-playable characters in a game what to do. Scientists can use AI to predict the weather or to analyze tons of pictures of space and find a new planet. We also have AI tools that regular people can use for fun or to learn things. There are AI tools called chatbots that you can type back and forth with like you’re talking to a real person. People use chatbots to answer all sorts of questions, from “Where is that Pikachu toy that I ordered?” to “How are marshmallows made?”
Some robots in movies and TV shows are evil, so you might be worried that AI is evil, too. The truth is that you don’t need to worry about intelligent robots taking over the world, but that doesn’t mean that you should always trust AI like you’d trust a parent or friend. AI can make mistakes just like a person can, so it’s not always right about everything. It can also have unfair biases: If the data that the AI was trained with included unfair statements about certain groups of people, the AI will think the same way. People can also use AI on purpose to make things that look real but aren’t. This can be fun, like making a picture of a purple unicorn that looks like a photo someone took in real life, but it can also be dangerous, like making a video of the president saying something that they never said. You should also know that AI can’t always keep a secret. Just like you shouldn’t share personal information online, you shouldn’t share it with an AI tool.
Using AI is a lot like using any other powerful tool, like a computer, a drill, or a car. We can do a lot of really cool and helpful things with AI, but we always need to remember to be careful with how we use it and how much we trust what it tells us.