Why is context stuffing the opposite of context engineering, even on a 1M-token window?

Context engineering is the discipline of choosing what reaches the model's context window and how it is structured. The whole point is that context is a scarce resource even when the window is large. Dumping the entire knowledge base into context goes against this in three concrete ways. **Cost.** API providers charge per input token. A 1M-token prompt costs roughly 125 times more than an 8K-token prompt on the same model. If you run that on every query, your unit economics break. **Latency.** Time to first token scales with input length. A 1M-token prompt takes seconds before the model even starts generating. Users notice. **Accuracy.** The lost-in-the-middle effect is a real, measured phenomenon: long-context models attend strongly to the start and end of the input and skip content in the middle. Burying the right chunk in the middle of 999,000 other tokens makes it harder for the model to use, not easier. A large window gives you a higher ceiling for legitimate cases (long documents, long conversations). It is not permission to skip retrieval.

**Context engineering** treats the context window as a scarce resource and decides what reaches it. Stuffing inverts every choice that discipline makes. The first issue is cost. Input tokens are billed at the input rate, and that rate does not get a volume discount past 1M. A query that should have been 8K input tokens but runs at 1M is paying a 125x penalty on every single call. Prompt caching only amortizes the static portion; the dynamic chunks pay full price every turn. The second issue is latency. Time to first token grows roughly linearly with input length on every modern model. A 1M-token prefill is multiple seconds before the model emits a single output token. For interactive applications that is fatal. The third and most counter-intuitive issue is accuracy. The lost-in-the-middle effect, well documented by Liu et al. and replicated in every long-context eval since (RULER, BABILong, NoLiMa), shows that attention concentrates at the head and tail of long contexts. The middle is under-attended. Stuffing buries the right chunk in the middle of a million distractors and gives the model permission to spread attention thinly. Accuracy on retrieval and reasoning tasks drops measurably. The right framing is that a 1M-token window is a higher hard budget. The soft budget, the point past which more tokens stop helping, is usually a small fraction of that and is what good context engineering targets.