AI news is full of confusing words: LLM, tokens, hallucinations, and more. You don't need a computer science degree to understand them. This glossary explains the most common AI terms in plain English, with simple examples.
Artificial Intelligence (AI)
AI is software that can do tasks that normally need human thinking, like understanding language, recognizing images, or making predictions. For a full introduction, read What Is Artificial Intelligence and How Does It Work?
Machine Learning (ML)
Machine learning is a way of building AI where the computer learns patterns from examples instead of following fixed rules. For example, it learns what a cat looks like by seeing thousands of cat photos.
Model
A model is the "brain" that results after an AI system has learned from data. When people say "GPT-style model" or "Claude model," they mean this trained system.
Training Data
This is the huge collection of text, images, or other material that an AI learns from. The quality of the data affects the quality of the AI's answers.
LLM (Large Language Model)
An LLM is an AI model trained on very large amounts of text so it can understand and write language. Chatbots like ChatGPT, Gemini, and Claude are built on LLMs. They work by predicting what words are likely to come next, based on patterns they learned.
Prompt
A prompt is the instruction or question you give to an AI. A clear prompt gets a better answer. See our guide on how to write better AI prompts.
Token
A token is a small piece of text that AI reads and writes, such as a word, part of a word, or punctuation. AI tools count tokens to measure input and output size, and many services limit or charge by tokens. As a rough idea, a short word is often one token, while a long word can be several.
Context Window
The context window is how much text an AI can "keep in mind" at one time, including your messages and its replies. If a conversation gets too long, the AI may forget earlier parts.
Hallucination
A hallucination is when an AI gives an answer that sounds confident but is false or made up, like inventing a fact, a quote, or a source. This is why you should always check important information, such as numbers, names, and dates, with trusted sources.
Bias
AI can pick up unfair patterns from the data it learned from, which can show up in its answers. Being aware of this helps you question results instead of accepting them blindly.
Generative AI
Generative AI creates new content, such as text, images, audio, or video, instead of only analyzing existing data. Chatbots and AI image tools are examples.
Deepfake
A deepfake is a fake photo, video, or voice made with AI to look or sound like a real person. They can be used for scams or false news. Learn how to protect yourself in our guide on the 7 signs of an AI-powered scam.
Chatbot
A chatbot is a program you talk to in text or voice. Some follow simple scripts, while modern ones use LLMs to hold natural conversations.
Fine-Tuning
Fine-tuning means giving an existing AI model extra training on specific data so it gets better at a certain job, like answering medical questions or writing legal summaries.
AI Agent
An AI agent is an AI system that doesn't just answer questions but can take steps to complete a task, like searching, filling in information, or using other tools.
Quick Cheat Sheet
- LLM: AI trained on large text to understand and write language
- Prompt: your instruction to the AI
- Token: a small piece of text the AI reads or writes
- Hallucination: a confident but wrong or made-up answer
- Deepfake: an AI-made fake photo, video, or voice
Final Thoughts
Knowing these terms makes AI news and tools much less confusing. The most important idea to remember: AI is powerful, but it can make mistakes, so check anything important. Bookmark this glossary and return whenever you meet a new AI word.
