Home › Glossary

AI Glossary, in plain English

Every term you'll meet in reviews and news posts — explained without the jargon.

A

Agent

An AI system that can plan and take multi-step actions toward a goal — browsing, calling tools, writing files — rather than just answering once.

ASR (Automatic Speech Recognition)

Software that converts spoken audio into text. The engine behind dictation tools and meeting transcribers.

C

Context Window

How much text a model can consider at once — conversation history, documents, instructions. Bigger windows handle longer projects.

Chain-of-Thought

Prompting a model to reason step by step before answering, which improves accuracy on complex problems.

D

Diffusion Model

The technique behind most AI image generators: starting from noise and gradually refining it into an image that matches your prompt.

F

Fine-tuning

Additional training on your own examples to specialize a model's tone, format, or domain knowledge.

H

Hallucination

When an AI states something false with confidence. Why citations, sources, and human review still matter.

L

LLM (Large Language Model)

The core technology behind chatbots like ChatGPT and Claude — trained on vast text to predict and generate language.

M

Multimodal

Models that work across formats — reading images, hearing audio, watching video — not just text in, text out.

P

Prompt Engineering

Designing instructions that reliably get the output you want. Better prompts beat better models more often than people expect.

R

RAG (Retrieval-Augmented Generation)

Grounding a model's answers in your own documents by retrieving relevant passages first — the cure for hallucinated facts.

T

TTS (Text-To-Speech)

Generating natural-sounding speech from text. Modern TTS handles emotion, pacing, and voice cloning.

Token

The unit of text models read and bill by — roughly three-quarters of a word in English.