AI Glossary
Plain-English definitions for the jargon you hit reading AI tool docs and articles — no circular definitions, no unexplained acronyms. 34 terms and counting.
Core Concepts
LLM (Large Language Model)
The type of AI that powers ChatGPT, Claude, and Gemini — trained on huge amounts of text to predict and generate language.
▸ READ DEFINITIONToken
A chunk of text (roughly ¾ of a word) that AI models read and generate one piece at a time.
▸ READ DEFINITIONContext Window
The maximum amount of text an AI model can 'see' and remember at once during a conversation.
▸ READ DEFINITIONHallucination
When an AI confidently states something false or made-up as if it were fact.
▸ READ DEFINITIONRAG (Retrieval-Augmented Generation)
A technique where an AI looks up real documents or sources before answering, instead of relying only on what it memorized during training.
▸ READ DEFINITIONFine-Tuning
Further training an existing AI model on a specific, narrower dataset to specialize it for a particular task or style.
▸ READ DEFINITIONEmbeddings
A way of converting text (or images) into numbers that capture their meaning, so a computer can compare how similar two pieces of content are.
▸ READ DEFINITIONMultimodal AI
AI that can understand and work with more than one type of input — text, images, audio, or video — not just text.
▸ READ DEFINITIONAI Agent
An AI system that can take multiple steps and use tools on its own to complete a task, rather than just answering one question at a time.
▸ READ DEFINITIONKnowledge Cutoff
The date after which an AI model has no information, because its training data ends there.
▸ READ DEFINITIONParameters
The internal numerical values an AI model learns during training — roughly, a measure of the model's size and capacity.
▸ READ DEFINITIONInference
The process of an already-trained AI model actually generating a response to your input.
▸ READ DEFINITIONOpen-Weight Model
An AI model whose trained parameters are publicly downloadable, so anyone can run or modify it themselves.
▸ READ DEFINITIONChain-of-Thought Reasoning
When an AI model works through a problem in visible, step-by-step reasoning before giving a final answer.
▸ READ DEFINITIONVector Database
A database built to store and search embeddings — the numerical representations of meaning that let AI find similar content.
▸ READ DEFINITIONBenchmark
A standardized test used to measure and compare how well different AI models perform on specific tasks.
▸ READ DEFINITIONSmall Language Model (SLM)
A compact AI language model with far fewer parameters than flagship models — faster and cheaper, but less broadly capable.
▸ READ DEFINITIONUsing AI Tools
Prompt Engineering
The skill of writing instructions to an AI model that reliably get the output you actually want.
▸ READ DEFINITIONZero-Shot / Few-Shot
Whether you give an AI zero examples (zero-shot) or a few examples (few-shot) of what you want before asking it to do the task.
▸ READ DEFINITIONTemperature
A setting that controls how random or predictable an AI's responses are — low is focused, high is creative and varied.
▸ READ DEFINITIONSystem Prompt
Hidden background instructions that set an AI's behavior, tone, and rules before you type anything.
▸ READ DEFINITIONAPI (Application Programming Interface)
A way for developers to plug an AI model directly into their own app or website, instead of using the AI company's chat interface.
▸ READ DEFINITIONFunction Calling / Tool Use
The ability of an AI model to call external tools or APIs — like a calculator, search engine, or database — instead of just generating text.
▸ READ DEFINITIONJailbreak (Prompt Injection)
A crafted input designed to trick an AI model into ignoring its safety rules or instructions.
▸ READ DEFINITIONRate Limit
A cap on how many requests you can send to an AI tool or API within a given time period.
▸ READ DEFINITIONAPI Key
A unique password-like code that identifies you when your app or script sends requests to an AI model.
▸ READ DEFINITIONGuardrails
Rules and filters built around an AI system to keep its behavior safe, on-topic, and within intended limits.
▸ READ DEFINITIONHow AI Sees & Creates
Diffusion Model
The technique behind most AI image generators — it starts with random noise and gradually refines it into a coherent image matching your prompt.
▸ READ DEFINITIONUpscaling
Using AI to increase an image's resolution, adding believable detail rather than just stretching it.
▸ READ DEFINITIONText-to-Speech (TTS)
AI that converts written text into natural-sounding spoken audio.
▸ READ DEFINITIONVoice Cloning
AI that recreates a specific person's voice from a short audio sample, so it can generate new speech in that voice.
▸ READ DEFINITIONInpainting
Using AI to fill in, remove, or replace a specific selected area of an image while leaving the rest untouched.
▸ READ DEFINITIONSeed
A starting number that controls the randomness of an AI image generation, letting you reproduce or slightly vary the same result.
▸ READ DEFINITIONNegative Prompt
A list of things you explicitly tell an AI image generator to avoid including in the result.
▸ READ DEFINITION