AI glossary, no jargon
Every AI term you'll run into, defined in one or two plain sentences.
All glossary terms
AI Agent
An AI agent is an LLM that plans and takes multi-step actions with tools — booking, browsing, coding — not just chatting. Plain-language definition.
Computer Vision
Computer vision is AI that understands images and video — how it spots tumors, reads X-rays and powers your phone's camera. Plain-language definition.
Context Window
A context window is how much text an AI can consider at once — its short-term memory. What it means, why it matters and how big they have gotten.
Embeddings
Embeddings turn words and documents into lists of numbers so machines can compare meaning. Plain-language definition and why search uses them.
Explainability
Explainability means an AI's decision can be explained in terms a human can check. Why regulators demand it and what "explainable AI" really means.
Fine-tuning
Fine-tuning retrains an AI model on your examples to specialize its behavior. When it is worth it, and when RAG is the better fix.
Hallucination
An AI hallucination is a confident but wrong answer from an AI model. Why it happens and how to catch it, in plain language.
LLM
LLM stands for large language model — the AI behind ChatGPT and Claude. One-line definition plus how it works, in plain language.
Multimodal
Multimodal AI works across text, images, audio and video in one model. What that unlocks and why it became the default.
Parameters
Parameters are the internal values a model learns during training — the numbers behind billion-parameter models. What they are and why size is not everything.
Prompt Engineering
Prompt engineering is the skill of writing instructions that get AI to produce what you actually want. The craft, minus the hype.
RAG
RAG (Retrieval-Augmented Generation) means an AI looks up facts before answering. Short definition, why it matters, and where you meet it.
Token
AI models read text as tokens — chunks of a few characters. Why tokens matter for pricing, limits and why models miscount letters.
Vector Database
A vector database stores embeddings and finds similar meanings fast — the search engine behind RAG and semantic search, explained simply.