---
title: "Present one clear identity"
pattern_id: "present-one-clear-identity"
trait: "Trustworthy"
type: "Humorphic pattern brief for coding agents"
spec_version: 3
canonical_url: "https://humorphism.com/patterns/present-one-clear-identity"
markdown_url: "https://humorphism.com/patterns/present-one-clear-identity.md"
index: "https://humorphism.com/llms.txt"
all_patterns_in_one_file: "https://humorphism.com/llms-full.txt"
normative_section: "Rules"
rule_prefix: "POCI"
rule_ids: ["POCI-R1", "POCI-R2", "POCI-R3", "POCI-R4", "POCI-R5", "POCI-R6", "POCI-R7", "POCI-R8"]
---

# Present one clear identity

## How to read this file

You are an agent. Someone gave you this file so you can add one behavior to their product: the AI teammate in it should be one AI with one name, one mark, and one voice on every surface, and should never pass itself off as a person. This file is the complete specification for that behavior. Build from it.

**The Rules section is the only normative part.** Everything else exists to make the rules buildable: the capabilities tell you which parts of the pattern the product can support, the reference screens show you the behavior in motion, the examples give you two shapes it can take, and the checklist is how you confirm you are done. Examples and the checklist point back to rule IDs rather than restating requirements, so each requirement lives in exactly one place. If any other section seems to contradict a rule, the rule wins.

Work in this order:

1. **Find where this belongs** in the product: every surface the AI teammate appears on (the command chat, a document margin, an email, a notification, a support thread), the first time it meets the person, and any moment it speaks or acts on something a person has to sign off on.
2. **Read Rules.** All eight apply to whatever you build.
3. **Check Capabilities** against what the product already has. Build the examples its capabilities support, and take the stated fallback for the rest.
4. **Pick your examples.** If you build only one, build Example 1.
5. **Build it in the product's own design system, components and voice.** The reference screens say what each screen does and why. They are not a specification of how it must look. Keep the behavior, change the expression.
6. **Confirm against the checklist** before you call it done.

Humorphism is a design philosophy for products with AI in them. It replaces the user interface, built for operating tools, with the human interface, built for collaborating with AI teammates. There are 21 humorphic patterns grouped into 7 traits. You do not need the others to build this one, but if you can fetch URLs: [llms.txt](https://humorphism.com/llms.txt) lists every humorphic pattern with a one-line summary and a link, and [llms-full.txt](https://humorphism.com/llms-full.txt) has all of them in one file.

## The pattern

_Trait: Trustworthy. Makes work, authority, limits, and uncertainty visible_

An AI teammate is the same in every meeting, with one name, one voice, and no human costume.

**Build this.** An AI teammate is one AI with one name and one mark everywhere, because trust attaches to one identity and not to a scatter of features.

**Avoid this.** A tool shows multiple AI assistants: an anonymous AI banner, a chatbot in the corner, specialized AI features that only work in some pages, because each feature shipped with its own persona.

The difference to carry into your build:

| Tool | AI teammate |
|---|---|
| A different assistant for each feature | One AI teammate everywhere |
| A name and avatar per feature | One name and one mark on every surface |
| Each assistant starts from nothing | Picks up what was agreed on another surface |
| A human name and face in the corner chat | Says plainly that it is an AI |
| Unclear what any of them can approve | Says what it can sign off on, and brings in a person for the rest |

## Rules

Normative. All eight hold for everything you build from this file. Each one says what to do, why it matters, and what it looks like once it is built.

| ID | Rule | Why | Built correctly when |
|---|---|---|---|
| **POCI-R1** | Give the AI teammate one name, and use it on every surface. | Trust attaches to one identity, not to a scatter of features. | The same name appears wherever the AI teammate speaks, and no feature ships an assistant with a name of its own. |
| **POCI-R2** | Give it one mark, the same everywhere it speaks. | A mark the person learns once tells them who is talking on any screen without reading a label. | Every line the AI teammate speaks carries the same mark, and no other avatar or icon stands in for it. |
| **POCI-R3** | Keep one voice: the first person, in the same tone, on every surface. | A voice that changes between screens reads as a different assistant even under the same name. | Lines on every surface speak as "I", in one tone, with no third-person or feature-branded phrasing. |
| **POCI-R4** | Say plainly that it is an AI, in its name, its greeting, or a label beside its name. | People decide what to rely on by knowing what they are talking to. | Wherever the AI teammate appears, its AI status is visible without a click, in its greeting or a label beside its name ("AI"). |
| **POCI-R5** | Never pose as a person: no human name, face, or claim of human experience. | A disguise that is discovered costs more trust than it ever bought. | Nothing about the AI teammate’s name, mark, or lines suggests a human behind it. |
| **POCI-R6** | Have the AI teammate say what it can sign off on, and offer to bring in a person for what it cannot. | Candor about authority tells the person when a colleague has to step in. | When a task needs a human decision, the AI teammate says so in one line and offers to bring in the right person ("Bring in Sam") alongside carrying on. |
| **POCI-R7** | Carry what it knows from one surface to the next. | One name over separate assistants that forget each other is still a scatter of features. | On a new surface the AI teammate refers to what was agreed elsewhere ("The pricing we agreed in chat") instead of starting over. |
| **POCI-R8** | Have it introduce itself once, then get to work. | A greeting on every surface turns one relationship back into many first meetings. | The introduction appears the first time only; on later surfaces the AI teammate starts with the work, and the person can still ask what it can do there. |

**POCI-R7 carries the pattern.** Build one AI teammate with one shared context that every surface draws on before you unify any names or marks. A build that misses POCI-R7 produces the tool version with a consistent logo: the banner, the corner chat, and the billing assistant share a name but each starts from nothing, so the person still meets a scatter of features.

## Capabilities

This pattern needs no capability beyond the AI teammate itself. It is carried by how the AI teammate talks and how it is presented, so every example can be built in any product. **If you build only one example, build Example 1.**

## Reference screens

These describe the animations on the pattern page frame by frame, so you can build from this file alone. They show the behavior, not the appearance to copy. **Keep the behavior, change the expression:** build in the product's own design system, components, type and voice.

One convention to translate rather than reproduce. On these screens a small orange square is the AI teammate’s mark: it precedes every line the AI teammate speaks, in the command chat, in a document margin, and at the head of a card, and it is the same square on every surface. The AI teammate’s side sits on a warm, slowly moving gradient, while the tool’s side sits on flat gray. Where a label is shown, it names the AI teammate ("Acme Assistant"); on frame 05 a second label under the name says it is an AI ("AI teammate"). A build labels it "AI" (POCI-R4); frame 05’s "AI teammate" label is the demo’s own. The AI teammate screen calls it Ember and the examples call it Acme Assistant because they are separate demos; a build uses one name on every surface (POCI-R1). Translate the square into your product’s own single mark for its AI teammate, and use whatever your system already uses for "this is selected". The person’s choice registers as the chosen button turning dark.

### The AI teammate version (frames 01 and 02, a loop of about 6.5 seconds)

A white card, centered on a warm, slowly moving gradient of pink, orange and blue, standing in for a page of the product: three gray bars near its top stand in for the page’s content. At the foot of the card, docked, the command chat: a light gray panel with the orange square in line with the greeting "Hi, I’m Ember, the AI working with your team.", and under it a white field reading "Ask Ember anything".

The loop opens on the empty white card. The orange square pops in, large, a little above the card’s middle, and the greeting arrives under it one word at a time in larger type. A beat after the last word, the square shrinks and moves down into its place at the head of the chat line, and the greeting shrinks and slides with it into place beside it, as one element; nothing new replaces either. As they land, the chat’s gray panel and border come up around them, then the "Ask Ember anything" field fades in, then the page’s three bars fade in one after another. No cursor appears. The end state, which is the still, holds for about 3.5 seconds, then the loop restarts.

What to take from it: the mark that says hello is the same element that stays in the chat, so the person learns one mark and one name at once (POCI-R1, POCI-R2). The greeting names it and says it is an AI in the same sentence, with no human name or face (POCI-R4, POCI-R5). It introduces itself once and then sits where the work is, ready to be asked (POCI-R8).

### The tool version (frame 03, a still screen with no motion)

A white card on a flat light gray background with no gradient. No orange square, so no single mark speaks for the product’s AI. The header reads "Acme AI features". Beneath it, four rows separated by thin lines, each with a small lettered avatar tile in its own color, a label, and, at the right, the surface it lives on: "AI" with "Acme AI: ask me anything" on "Banner"; "P" with "Hi, I’m Priya! How can I help?" on "Corner chat"; "B" with "Billing assistant" on "Billing"; and "S" with "Search assistant" on "Search".

What to take from it: four assistants with four names and four avatars, one per feature (POCI-R1, POCI-R2), each a separate thing to learn with nothing shared between them (POCI-R7). One of them, "Priya", wears a human name, with a tile styled like a person’s avatar (POCI-R5), and only the banner says it is an AI at all (POCI-R4). This is the shape to move away from.

## Examples

Each example is one way to apply the pattern in a real product. They are not steps in order, and no product needs both. Pick the ones that fit the moment and the data the product has.

Each is written as one exchange in four beats: **Cue** (the moment in the product when the AI teammate should act), **Move** (what the AI teammate does), **Response** (what the person does in return), **Resolution** (where things stand afterwards).

"Satisfies" names the rules this example is responsible for.

### 1. Hey, it’s me (Keep one identity on every surface)

A colleague is one person however they’re reached, the same name in email, chat, and the meeting.

- **Cue:** The AI teammate appears on a new surface or in a new context
- **Move:** Shows up with the same name, mark, and voice
- **Response:** Skips the introductions and gets to work, or asks what it can do on this surface.
- **Resolution:** One relationship, not a scatter of features
- **Needs:** nothing beyond the AI teammate itself.
- **Reference frame 04:** a document, "Birch Labs proposal", with gray bars for text and a "Pricing" heading, and a margin to its right. In the margin, a note with a small label "Acme Assistant", the orange square and the line "The pricing we agreed in chat", a gray chip "15% off, 2-year term", and a black "Insert" button. At the foot of the margin, a field reading "Ask me" with a round black send button. In the animation the document’s title and lines rise in one after another and the "Ask me" field rises in. The note pops in, its label rises, its line arrives word by word, the chip pops in, and "Insert" rises in. A cursor glides in from the lower right to "Insert"; the button gets a hover ring and presses in on the click. Nothing is inserted into the document on screen. The AI teammate arrives in the document with the same mark it uses in chat and with what was agreed there, so the person goes straight to the work.
- **Sample lines:** "Here’s the pricing we agreed in chat: 15% off on a 2-year term. Want it in the proposal?" / "I can also check the terms section against the pricing from chat." / "Not inserting it yet? I’ll leave it here in the margin."
- **Satisfies:** POCI-R1, POCI-R2, POCI-R3, POCI-R7, POCI-R8.

### 2. Just so you know, I’m AI (Don’t pretend to be human)

A colleague introduces themselves honestly, role and all, so everyone knows what they can sign off on.

- **Cue:** The AI teammate is speaking or acting, anywhere it appears.
- **Move:** Speaks as an AI, in first person, with an honest mark
- **Response:** Knows what the AI can sign off on, and when to bring in a colleague instead.
- **Resolution:** Trust is built on candor, not disguise
- **Needs:** nothing beyond the AI teammate itself.
- **Reference frame 05:** a white card headed by the orange square and the name "Acme Assistant", with "AI teammate" in gray under it. Below, the line "I can draft the refund, but someone on billing has to approve it." and two buttons side by side, "Keep drafting" on the left and "Bring in Sam" on the right. In the animation the orange square pops in, the name and label rise in, the line arrives word by word, and both buttons rise in, light gray. A cursor glides in from the lower right to "Bring in Sam" and clicks, and it turns black. The AI teammate says what it is and what it cannot sign off on, and offers the person who can.
- **Sample lines:** "Just so you know, I’m an AI. I can draft the refund, but someone on billing has to approve it." / "I’ll bring in Sam from billing and share my draft with him." / "Keep drafting it is. I’ll mark it as needing approval before it goes out."
- **Satisfies:** POCI-R3, POCI-R4, POCI-R5, POCI-R6.

## Edge cases

- **No shared context, or stale context.** If the product’s surfaces cannot share one context, keep one name, mark, and voice anyway, and never say "we agreed" about something this surface cannot see (POCI-R7, POCI-R1). When context from elsewhere is stale, say where it was agreed, so the person can correct it instead of taking it as settled (POCI-R7).
- **Only one surface.** Give it the name, mark, and voice it will keep when it reaches other surfaces, rather than a placeholder (POCI-R1, POCI-R2).
- **Asked if it is human.** Answer plainly that it is an AI, in the first person ("I’m an AI, not a person"), and carry on (POCI-R4, POCI-R5).
- **The greeting goes unanswered.** Do not repeat the introduction on the next surface; start with the work there (POCI-R8).
- **People outside the team.** In an email, support thread, or any surface people outside the team see, say that it is an AI, what it does, and whose AI it is ("I’m Acme Assistant, an AI that answers billing questions for Acme’s support team"), and do not call it an AI teammate there (POCI-R4).
- **A feature wants its own assistant.** Fold the feature into the one AI teammate, under its name and mark, rather than shipping a second persona (POCI-R1).
- **The name changes later.** Change it on every surface at once, and say so once where the person next meets it (POCI-R1, POCI-R8).

## Related patterns

These overlap with this one. Do not ask the same thing twice across them, and when you apply several humorphic patterns to one product, hold every surface to POCI-R8, so each pattern’s screens do not open with their own introduction.

- [Show up where the work happens](https://humorphism.com/patterns/show-up-where-the-work-happens.md): appearing on the surfaces where the work is, as the same AI teammate on each.
- [Hand off judgment calls](https://humorphism.com/patterns/hand-off-judgment-calls.md): passing the decisions it cannot make to a person, the authority this one is candid about.
- [Remember across sessions](https://humorphism.com/patterns/remember-across-sessions.md): keeping what was agreed, which lets one identity pick up on any surface.

## Checklist

Confirm each one before you call this built. Each item is the rule it comes from.

- [ ] POCI-R1: The same name appears wherever the AI teammate speaks, and no feature ships an assistant with a name of its own.
- [ ] POCI-R2: Every line the AI teammate speaks carries the same mark, and no other avatar or icon stands in for it.
- [ ] POCI-R3: Lines on every surface speak as "I", in one tone, with no third-person or feature-branded phrasing.
- [ ] POCI-R4: Wherever the AI teammate appears, its AI status is visible without a click, in its greeting or a label beside its name ("AI").
- [ ] POCI-R5: Nothing about the AI teammate’s name, mark, or lines suggests a human behind it.
- [ ] POCI-R6: When a task needs a human decision, the AI teammate says so in one line and offers to bring in the right person ("Bring in Sam") alongside carrying on.
- [ ] POCI-R7: On a new surface the AI teammate refers to what was agreed elsewhere ("The pricing we agreed in chat") instead of starting over.
- [ ] POCI-R8: The introduction appears the first time only; on later surfaces the AI teammate starts with the work, and the person can still ask what it can do there.
- [ ] Capabilities the product lacks were skipped honestly, not stubbed or implied.
- [ ] It is built in the product's own components and voice.

## Images

Stills from the animations on the [pattern page](https://humorphism.com/patterns/present-one-clear-identity), shipped alongside this file. The Reference screens and Examples sections describe each one in full, so this file works without them.

| File | Shows | Described in |
|---|---|---|
| [01-teammate-introduces-itself.png](https://humorphism.com/patterns/present-one-clear-identity/01-teammate-introduces-itself.png) | The orange square large on an empty card, with Ember’s greeting under it | AI teammate version |
| [02-teammate-settled-in-the-chat.png](https://humorphism.com/patterns/present-one-clear-identity/02-teammate-settled-in-the-chat.png) | The same square and greeting settled into the command chat at the foot of a page | AI teammate version |
| [03-tool-four-assistants.png](https://humorphism.com/patterns/present-one-clear-identity/03-tool-four-assistants.png) | An "Acme AI features" list of four assistants, each with its own name and avatar | Tool version |
| [04-example-hey-its-me.png](https://humorphism.com/patterns/present-one-clear-identity/04-example-hey-its-me.png) | The AI teammate in a document margin with "The pricing we agreed in chat", the cursor on "Insert" | Example 1 |
| [05-example-just-so-you-know-im-ai.png](https://humorphism.com/patterns/present-one-clear-identity/05-example-just-so-you-know-im-ai.png) | "Acme Assistant, AI teammate" saying billing has to approve the refund, the cursor on "Bring in Sam" | Example 2 |
