Module 03Free
The context window, why long chats lose detail, and why a new chat is a stranger.
Module goal
Understand what the AI can currently see, and learn to restate key facts during long working sessions.
An AI tool doesn't actually “remember” you, your business or your previous conversations by default, unless a product is specifically built to do that. Within a single conversation it can see everything you and it have typed so far — this is the context window, and it works like a working memory. Everything currently in that window shapes the next response.
Every AI tool can only hold a certain amount of text in working memory at once. If a conversation or document gets long enough, older parts fall out of view — the AI genuinely cannot see them anymore, even though you can still scroll up and read them.
Real-life example: Yaw, a repair shop owner, pastes a year of customer complaint records into a chat and works through fifty messages of drafting and pattern-spotting. Then he asks, “What was that first complaint about a cracked screen from January again?” That January detail may already have slipped out of the AI's working memory. It isn't being careless — it can no longer see it.
Unless you are using a tool with a specific memory feature, starting a fresh chat means the AI has no idea what you discussed yesterday, last week, or five minutes ago in a different window.
Real-life example: Adjoa spends Monday teaching an AI chatbot all about her catering business — menu, prices, delivery zones. On Tuesday she opens a new chat and asks, “What's my delivery fee for East Legon again?” That chat is a total stranger meeting her for the first time.
Practical implication
For long or complex work, periodically re-state the key facts you need the AI to keep in mind rather than assuming it still remembers them.
Question 1 of 5
What is the “context window”?