There's a conflation in how most people talk about AI memory that leads them to build the wrong thing. The context window and memory are not the same thing. They serve different purposes, operate on different timescales, and fail in different ways. Conflating them produces a system that looks like it handles memory but doesn't actually compound over time. Here's the distinction and why it matters practically. The context window The context window is the model's working memory — the span of tokens it can attend to in a single inference pass. Everything outside the window is invisible. Everything inside the window is equally weighted (with some positional decay depending on architecture).…