Context and context windows
Also used: Context, Context window, Context limit
Context is the information available to an AI model for the current response; the context window is the limited amount of information the model can consider at one time.
Why it matters to educators
Educators often work with long documents, evolving conversations, and multiple constraints. Knowing that context is limited helps explain why an AI tool may overlook an earlier direction or produce a weak answer from an incomplete source.
The foundation
The context can include the current prompt, earlier messages, uploaded material, application instructions, and information returned by connected tools. The context window is the capacity that holds this information while the model generates its response.
A larger context window can accommodate more material, but it does not guarantee that every detail will be used correctly. Context is also different from long-term memory: an application may save preferences or conversation history using a separate feature with its own privacy implications.
What this can look like in education
Align a lesson to several sources
A curriculum leader supplies the relevant standard, lesson draft, and review criteria together, clearly labeling each source and asking the tool to cite which source supports each recommendation.
Reset a drifting conversation
When a long planning chat starts ignoring constraints, the educator begins a focused request that restates the objective, non-negotiables, and current draft instead of relying on the entire conversation history.
