What Is AI?

AI, short for artificial intelligence, is technology that lets computers do things that normally require human intelligence, such as understanding language, recognizing faces, making recommendations and writing text. You probably use it every day: when your email filters out spam, your phone recognizes your face, a map predicts traffic, or a streaming service suggests what to watch next.

What artificial intelligence means

Artificial intelligence is a broad term for computer systems that work out, from the information they receive, how to produce outputs such as predictions, recommendations, decisions or new content. The OECD, in a definition updated in 2023, describes an AI system as a machine-based system that "infers, from the input it receives, how to generate outputs such as predictions, content, recommendations, or decisions."

Most AI today works by learning patterns from large amounts of data rather than following rules a programmer wrote for every situation. That approach is called machine learning. The short version of the definition lives in the LATHIC Glossary entry for artificial intelligence.

The easiest way to understand AI, though, is to look at where it already shows up.

Examples of AI you already use

A closer look at a few examples

Spam filters. Instead of relying only on a fixed list of banned words, modern spam filters learn from huge numbers of messages what spam and phishing tend to look like. Google says Gmail blocks more than 99.9% of spam, phishing attempts and malware from reaching users, and that its AI-enhanced filters block nearly 10 million spam emails every minute.

Face recognition on phones. Apple says Face ID uses the iPhone's TrueDepth camera and machine learning, and that the probability a random person could use their face to get into your iPhone is less than 1 in 1,000,000.

Streaming recommendations. Netflix says its recommendation system considers your viewing history and how you rated titles, what members with similar tastes watch, and details about titles such as genre and actors. It says it does not use demographic information such as age or gender to make those decisions.

Arrival times. DeepMind, now Google DeepMind, reported in 2020 that machine learning improved the accuracy of Google Maps' real-time arrival estimates by up to 50% in some cities.

Chatbots. Tools like ChatGPT are built on large language models that learned patterns from vast amounts of text. They are among the most visible kinds of AI today: a Pew Research Center survey conducted in February 2026 found that about half of U.S. adults use AI chatbots.

Robotaxis. As of September 2026, Waymo operates fully autonomous rides in a growing list of U.S. cities, including Phoenix, Los Angeles and the San Francisco Bay Area, and says it has served more than 20 million rides.

You use AI more than you might think

Many people may not notice how often they use AI, because it is built quietly into familiar products. In Pew Research Center surveys published in April 2025, 27% of U.S. adults said they interact with AI almost constantly or several times a day, while 79% of AI experts said people in the U.S. interact with AI that often.

A useful rule of thumb: when a product predicts something, recognizes something, personalizes something for you or generates something new, AI is often involved.

These everyday examples are only part of the picture. For how AI is applied in fields such as healthcare, finance, science and education, see What Is AI Used For?.

What does AI stand for?

AI stands for artificial intelligence. The term first appeared in 1955, in a proposal for a summer research workshop at Dartmouth College, and it has named the field of research ever since. "Artificial" means made by people rather than occurring naturally; "intelligence" refers to abilities such as learning, reasoning, understanding language and solving problems.

Narrow AI vs general AI

Every example in this article is narrow AI: AI designed for a specific task or a limited set of tasks. A spam filter cannot drive a car, and a navigation app cannot recognize your face.

Artificial general intelligence, or AGI, refers to a hypothetical AI that equals or surpasses human performance on all or almost all cognitive tasks. It is not part of the AI people use today. The International AI Safety Report 2025, prepared by an international panel of experts, describes AGI as a "potential future AI" and notes that the term "lacks a universal definition."

Chatbots like ChatGPT blur the line a little, because they can help with many different tasks, from writing to coding to planning. They remain uneven, though: the International AI Safety Report 2026 notes that today's general-purpose AI can excel at some difficult tasks while failing at some simpler ones. For more, see the glossary entries for narrow AI and artificial general intelligence.

AI vs automation

Automation means using technology to perform tasks with little or no manual effort, usually by following fixed rules. Modern AI usually means systems that learn from data to make predictions or decisions in situations that were not spelled out in advance. The two often work together, but they are not the same.

When AI is used to make automated systems more flexible, people call it AI automation. The LATHIC Glossary covers automation and how rule-based and AI-driven approaches differ.

How AI does all of this

Most of these examples work the same basic way. The AI is trained on large amounts of examples, finds patterns in them, and then uses those patterns to make predictions about new information: this email is probably spam, this face matches the owner, this driver will probably arrive in 18 minutes.

That is also why AI can be wrong. It makes predictions based on patterns, not certainties, and it can fail when something is unlike what it learned from. How Does AI Work? walks through the process step by step, and What Is Machine Learning? explains how the learning happens.

How to explain AI to someone new to it

If you are explaining AI to a relative, friend or colleague who is new to the idea, these approaches help:

• Start with something they already use. "You know how your phone suggests the next word when you type? That's AI."
• Use one simple idea. "AI is software that learns from lots of examples, then makes good guesses about new situations."
• Show, don't describe. Ask a voice assistant or chatbot a question together.
• Be honest about limits. Explain that AI can be very useful and also confidently wrong, so it is worth double-checking important answers.
• Skip the jargon. Terms like "neural network" and "large language model" can come later, if they are curious.