A context window is the amount of text an AI model can keep in view at one time. It covers everything the model is working with: its instructions, your conversation so far and any documents you paste in. The size is measured in tokens. Think of it like a desk. You can only spread out so many pages, and when the desk is full, older pages get pushed off or squeezed into a short summary.
Why it matters
Larger context windows let staff paste whole reports, donor lists or case files into a large language model in one go. That is convenient, but it also raises the stakes of what goes in. Before anyone pastes a full document, ask whether it holds personal or confidential information that should not be shared with an outside tool. Also keep in mind that in a long chat, early material can be dropped or summarized, so the model may lose track of instructions you gave at the start. When you talk to a vendor, ask what happens to the text you paste, how long it is kept, and whether the tool uses retrieval-augmented generation to bring in documents instead of having you paste them.
Published Updated