AI glossary

Plain-language definitions for the terms that matter in practice. Every entry links to the lesson that teaches it.

• Action item
  A single commitment from a meeting: one action, one owner, one date — or an explicit 'unassigned' / 'no date' where the…
• Agent
  A system that runs a model in a loop: given a goal, it decides the next step, calls tools, observes the result and…
• Aspect ratio
  The proportion between an image's width and height — 1:1 square, 16:9 widescreen, 4:5 portrait, and so on.
• Caveat
  A condition or limit attached to a statement — the 'if', 'unless' or 'to be confirmed' that decides when it is true.
• Context window
  The context window is everything a model can consider at one time: your current message, the conversation so far…
• Data classification
  Sorting content before it travels: public, internal, confidential, personal, secret.
• Diffusion
  A diffusion model generates an image by starting from random noise and refining it step by step toward something that…
• Dry run
  Running a script or process in a mode that only reports what it would do, without changing anything.
• Few-shot prompting
  Few-shot prompting means showing the model two or three input → output examples of the pattern you want, then giving it…
• Function calling
  A mechanism where the model, instead of inventing an answer, asks the surrounding system to run a named tool — check a…
• Generative AI
  Generative AI is software that produces new content — text, images, audio or code — by continuing patterns learned in…
• Grounding
  Grounding means constraining a model's answer to material you supply — a pasted policy, paper or document — instead of…
• Hallucination
  A hallucination is a fluent, confident AI answer — or a detail inside one — that is not true.
• Inference
  Running a trained model to get an output — every chat message, summary and image render is inference.
• Inpaint
  Editing only part of an image: you mark a region (a mask) and describe what should replace what is there.
• Large language model (LLM)
  A large language model is an AI system trained on enormous amounts of text to predict the next token in a sequence.
• Multimodal
  Working with more than one kind of input or output — text, images, audio, documents or video — in the same model.
• Negative prompt
  A dedicated field some tools offer for listing what to steer away from.
• Placeholder
  A marked gap in a draft — like [refund reference] or [amount] — that only you can fill with a real fact.
• Primary source
  The original artifact a claim comes from — the study itself, the law's text, the official documentation, the dataset.
• Prompt
  A prompt is everything you give a model to work with: the instruction, the context, the examples and any material you…
• Prompt injection
  Instructions hidden inside content you ask an AI to process — a document, web page or email — aimed at the assistant…
• Prompt template
  A prompt with named slots for the parts that change each time (topic, audience, source text) and fixed text for the…
• Reasoning model
  A reasoning model spends extra computation before answering — working through steps, checking itself, exploring…
• Redaction
  Removing or replacing sensitive values — names, keys, amounts — with safe stand-ins before text travels anywhere it…
• Rubric
  A short list of checkable questions you run against a draft before shipping it — is the format right, can every claim be…
• Seed
  A number that fixes the starting point of a generation's randomness.
• Structured output
  Output shaped to a fixed format — a table, a list of fields or JSON — so it can be reused by other tools instead of…
• Style reference
  An image (or its written description) used to guide the look of a new image — palette, light, texture, mood.
• Temperature
  A sampling setting that controls how randomly a model picks its next token: lower values make output more predictable…
• Token
  A token is a chunk of text — roughly a word piece — that a model processes as one unit.
• Training cutoff
  The date after which a model's training data stops.
• Triage
  Sorting incoming items — messages, tickets, requests — into defined categories and a priority order before working on…
• Unsupported claim
  A statement specific enough to be checked and consequential enough to matter — attached to no evidence.
• Workflow
  A fixed sequence you design — inputs, actions, outputs and checks — with AI steps placed inside a structure you control.