# The four properties of the machine
> **What you will learn**
> This part runs to more than ten chapters. Here is the map first: those
> chapters are each digging into one of **four properties**. And you will see
> why memorising a strengths list and a limits list separately is wasted effort.
## Do not memorise two lists
People learning AI are usually handed two lists like these.
```
Good at Bad at
───────────── ─────────────
Summarising Recent information
Translating Arithmetic
Drafting Checking facts
Classifying Remembering long documents
```
**Memorising these two lists buys you nothing.** Meet a task that is on neither
list and you cannot judge it; change the model and the lists change too.
What is useful sits underneath. **The left and the right are not different
facts.** They are one property seen from two sides.
> **The reason it translates well and the reason it cannot do arithmetic are
> the same reason.**
> Both follow from it being a machine that continues text.
## The four properties
| Property | What it makes possible | What it rules out, for the same reason |
|---|---|---|
| **1. Continuation** | Writes naturally in any register | Produces the shape of an answer while knowing nothing |
| **2. Learned knowledge** | Knows context you never mentioned | Knows nothing after its cutoff, nothing about your company |
| **3. Working memory** | Handles material you paste in, right there | Anything past the edge disappears silently |
| **4. Instruction response** | You change behaviour with words | What you meant and what landed drift apart |
In each row, **the left and the right are the first and second half of one
sentence.** Remove the right-hand side and the left goes with it.
```mermaid
graph TD
A["One property"] --> B["The capability side"]
A --> C["The limit side"]
B --> D["e.g. writes in any register"]
C --> E["e.g. cannot do arithmetic"]
D --> F["Both come from one thing —
continuing the next words"]
E --> F
```
### 1. Continuation — it does not find an answer, it continues one
It is not a search box. It is **an extremely sophisticated autocomplete.** Given
a question it does not pull an answer out of a store; it lays down, piece by
piece, the words likely to follow that question.
- In depth → [What a language model actually does](/guide/ai-language-model)
- The limit side → [Why it is wrong so plausibly](/guide/ai-hallucination)
### 2. Learned knowledge — only what it learned, only up to when it learned
Everything inside the model came from its training data. That leaves two kinds
of hole: **a hole in time** (anything after training ended) and **a hole in
scope** (your company's material, which was never public in the first place).
Both holes are filled the same way — **by giving it the material.**
- How to fill them → [Finding and attaching sources](/guide/ai-grounding)
- Which method to pick → [Which kind of learning?](/guide/ai-rag-vs-finetune)
### 3. Working memory — there is a window, and it has an edge
There is a fixed amount it can look at in one go. That means **the more material
you put in, the better it answers**, and it also means **when it overflows, the
front falls off.**
The falling-off happens without warning. The model does not say "I forgot"; it
just fills in.
- What the window is → [The context window](/guide/ai-context-window)
- When it overflows → [When context gets pushed out](/guide/ai-context-overflow)
- Keeping things outside the window → [Memory that outlives the session](/guide/ai-memory)
### 4. Instruction response — it follows, but not by understanding
Tell it "no tables" and it stops using tables. Then a table reappears in the
third answer. That is because it is **not keeping a rule it understood as
meaning; it is producing the continuation of a text that contains a rule.**
- In depth → [Why instructions do not simply stick](/guide/ai-steerability)
- Writing them well → [The four parts of a good instruction](/guide/ai-intent-context)
## When properties collide — a failure is rarely just one
Incidents in the field are seldom explained by a single property. **Two of them
overlap.**
```
Incident: asked to review a contract, and about thirty minutes in it
confidently cited a clause number that does not exist.
Property 3 (working memory): the front of the contract fell out of the window
Property 1 (continuation): the vacated slot got filled with "the shape of a
clause number"
```
**Falling off alone does not cause an incident.** It becomes one because
continuation fills the gap. So the response is also two-part — manage the window
(restate the constraints) and demand grounding (make it quote the clause text).
| Symptom | Properties colliding | Look at first |
|---|---|---|
| Cites a source that does not exist | 2 + 1 | Did you supply material? |
| Rules break down as the chat gets long | 3 + 4 | Did you restate the constraints? |
| Just keeps agreeing with you | 4 + 1 | Did you ask for the counter-case? |
| States a current figure confidently | 2 + 1 | Did you attach a tool? |
**Diagnosis does not start with "why was it wrong?" but with "which properties
collided?"** Name the property and the response follows.
## Can these properties be removed?
No. **They can be reduced.**
```
✗ "A better model will fix it"
→ The frequency drops; the property does not. Rarer errors are harder
to catch, not easier.
✓ "Which property does this task run into? What do I put in that slot?"
→ Supply material (2), manage the window (3), restate constraints (4),
demand grounding (1).
```
Every remaining chapter in this part is about **what you put in that slot.**
---
## Check
**1. Why is memorising a strengths list and a limits list separately wasted effort?**
Answer
**Because they are two sides of the same property.** The reason it translates
well and the reason it cannot do arithmetic both come from it being a
continuation machine. Knowing the properties lets you judge tasks that appear on
neither list; knowing only the lists leaves you stuck on anything new.
**2. What does "falling off alone does not cause an incident" mean?**
Answer
Material leaving the window is just information going missing. It becomes an
incident **because continuation fills the empty slot with something plausible.**
That is why the response has two halves: manage the window, and demand grounding.
**3. Do these four properties disappear as models improve?**
Answer
No. They come from the structure, so **only the frequency drops.** And as the
frequency drops people relax, which can make the remaining errors harder to
catch.
---
Now the first property in detail →
[What a language model actually does](/guide/ai-language-model)