The AI glossary for beginners

AI has its own vocabulary. Here are 12 essential terms explained simply so you can find your way.

• Generative AI : an AI that creates content (text, image, code) from a request.
• LLM (large language model) : the engine behind assistants like ChatGPT or Claude, trained on huge amounts of text.
• Prompt : the request you write to the AI. The better it’s phrased, the better the answer.
• Token : the basic unit the AI reads: a chunk of a word. Models count in tokens.
• Context (context window) : how much text the AI can “keep in mind” at once.
• Hallucination : when the AI confidently invents false information. Hence the need to verify.
• Multimodal : a model that understands several formats: text, image, audio, video.
• Agent : an AI that doesn’t just answer, but performs tasks (search, click, organize).
• Fine-tuning : retraining a model on specific data to adapt it to a use case.
• Open-source / open weights : a freely downloadable model you can host yourself.
• RAG : a technique that lets the AI look up information in your documents before answering.
• API : the technical “socket” that lets you plug an AI into an app or website.

Keep this page handy: each term will come up often in our articles.