# The AUX Manifesto

> Canonical: https://auxfirst.com/#manifesto
> License: CC BY 4.0 · auxfirst agency 2026

## From interface design to relationship design

For forty years, we perfected the art of making people tap screens: fewer clicks, clearer labels, smoother flows. We called it **user experience**, and it won markets.

But the day software gained memory, initiative, and judgment, the rulebook changed. A product with memory isn't a tool — it's a **collaborator**. And collaborators don't live in workflows. They exist within the context of relationships.

We call this shift **Agentic User Experience (AUX)** — the discipline of designing enduring, adaptive relationships between users and systems with memory, initiative, and judgment. Software that becomes more useful, more personal, and more trusted with every interaction.

Classic UX designs for discrete tasks. AUX designs for relationships.

> **Trust isn't a feature in this world. Trust is the product.**

## The defining shift

| Classic UX | AUX |
|---|---|
| Unit of design = a session | Unit of design = a relationship |
| Personalisation is a setting | Personalisation is the substrate |
| Users operate the tool | Humans supervise the collaborator |
| Errors cost a click | Errors cost trust |
| Better model = better product | Better trust layer = better product |

## The AUX Evolution Curve

A four-stage maturity ladder. Most teams shipping AI features today are at Stage 01 or 02. The strategic moats start at Stage 03.

### Stage 01 — Conversational
Natural language replaces rigid forms. The product can hold a real conversation. Still a tool — the user does all the steering. *Most AI features today.*

### Stage 02 — Task-Aware
The product can *do* things on the user's behalf. Watches interactions, adapts in real time. Still session-bound and mostly reactive. *Copilots in 2026.*

### Stage 03 — Personally Intelligent
Remembers preferences, tone, goals, and recurring tasks across sessions. Nudges. Anticipates. Feels like a colleague, not a feature. *Where the moat begins.*

### Stage 04 — Socially Embedded
Understands your team, role, shared vocabulary, and political reality. Not just personalised — part of how your organisation works. *Almost nobody is here yet.*

The further right you move on this curve, **the harder it gets to swap you out**. A Stage 01 chatbot is replaceable in an afternoon. A Stage 04 system with six months of context on your team is functionally irreplaceable. That's the AUX moat — built from time, trust, and accumulated context. It cannot be cloned by a competitor with more capital.

## The three properties that make something an agent

Subtract any one and you're back to building software.

- **Memory** — recall across sessions, transparent and editable
- **Initiative** — the system can propose and act, not just respond
- **Judgment** — the system can choose between options under uncertainty

## Where AUX sits in the stack

- **AUX (Agentic User Experience)** — the design layer. How humans and agents collaborate at the product surface.
- **AX (Agent Experience)** — the API layer. How agents and software systems negotiate underneath the surface.
- **TrustKit / AGT** — the enforcement layer. The deterministic policy code that bounds what the agent may do.
- **Agent Enablement** — the operational layer. The grounding, playbooks, tools, and feedback loops that make an agent reliable.

## The auxfirst position

We don't design interfaces. We design **relationships** between users and agents.

The agent is a user. The relationship is the product. Trust is the moat.

Welcome to AUX.

---

Related: [The six AUX patterns](https://auxfirst.com/patterns.md) · [The ten AUX heuristics](https://auxfirst.com/heuristics.md) · [The auxfirst services](https://auxfirst.com/services.md)