# Data masking

> **What this chapter covers**
> What gets hidden, why tokens rather than deletion, and why people can still see
> the original.

## Swapped on the way out

Switch on **Settings → Data masking** and it applies to **every message and
document excerpt** leaving this workspace.

![Data masking](/guide-assets/cn-settings-security.png)

```
Original:  Contact Kim 010-1234-5678 chulsoo@hanbit.co.kr
The model gets:  Contact Kim [phone#1] [email#1]
```

## Why tokens instead of deletion

Flattening to `[REDACTED]` **breaks the sentence so the model cannot read
relationships.**

```
✓ "[customer#1] supplies [customer#2]"   → the relationship survives
✗ "[REDACTED] supplies [REDACTED]"       → no information at all
```

And **the same value always gets the same token.** If "Acme" becomes
`[customer#1]` in one place and `[customer#2]` in another, the model reads two
different companies.

## Choose what to mask

| Item | How it is caught |
|---|---|
| Resident registration no. | Pattern |
| Card number | Pattern |
| Phone number | Pattern |
| Bank account | Pattern |
| Email | Pattern |
| **Glossary terms** | Words registered in the [glossary](/guide/cn-glossary-terms) |

**The last row is half of this feature.** Customer names, product codes and
internal names cannot be caught by patterns — they differ per workspace, so they
cannot live in code. The glossary is that list.

> If a phone pattern matched an ID number first, the tail would survive. So **more
> specific rules are checked first** — the order you tick the boxes doesn't matter.

## Use the preview

The **preview** at the bottom shows "what the model actually receives" and updates
as you tick items. **Look at it before switching this on.**

## People still see the original

The most common misreading.

> **Masking hides things from the model, not from people.**

The substitution dictionary is kept, so authorised users can be shown the original.
Hiding things from people is a job for
**[visibility and permissions](/guide/cn-security)** instead.

## Tier

A Business-tier feature ([Plans and tiers](/guide/cn-plans)). On lower tiers the
screen is locked — but what is locked is the policy screen, not your data.

## Where to switch it on

Everywhere is not the right answer.

| Workspace | Recommendation |
|---|---|
| Customer support, HR, legal | On |
| Anything with external intake (widget, email) | On |
| Internal engineering and planning | Usually unnecessary |

**Masking is not free.** The model never sees the real values, so work that has to
handle the values themselves — validating a phone number format, say — cannot be
done in a masked workspace. Split those into their own workspace.

## Common misunderstandings

### "Does masking make us compliant?"

**No.** It closes one path to the model. Lawful basis, retention periods and
deletion procedures all remain ([Regulation and compliance](/guide/ax-compliance)).

### "Is personal data inside documents masked too?"

**Drive excerpts take the same path.** The original file is untouched though, so
anyone who downloads it gets the raw text. That is a
[drive permissions](/guide/cn-drive) question.

---

## Check yourself

**1. Why tokens rather than deletion?**

<details>
<summary>Answer</summary>

**The sentence structure has to survive for the model to read relationships.**
"[customer#1] supplies [customer#2]" keeps the relationship; two `[REDACTED]`s
carry nothing.
</details>

**2. How do you mask a customer name?**

<details>
<summary>Answer</summary>

**Register it in the glossary and turn on "mask glossary terms".** Patterns cannot
catch it, and the value differs per workspace so it cannot live in code.
</details>

**3. Does masking hide things from teammates?**

<details>
<summary>Answer</summary>

**No.** It hides things from the model. Hiding from people is visibility and
permissions.
</details>

---

Next, what gets recorded → [Audit log](/guide/cn-audit)