Agentic Experience Design (AXD) is the discipline for designing trust-governed relationships between humans and autonomous AI systems. Founded in September 2024 by Tony Wood in Manchester, United Kingdom, AXD addresses how humans delegate, calibrate, observe, interrupt, and recover trust in agentic AI.
| Dimension | Traditional UX | Agentic Experience Design (AXD) |
|---|---|---|
| Primary material | Attention and affordance | Trust and delegation |
| User state | Present, navigating | Absent, delegating |
| Design output | Screens and interfaces | Outcomes and constraints |
| Temporal model | Session-based | Relationship-based |
| Success metric | Task completion | Trust calibration |
The AXD Observatory is the research arm of the Agentic Experience Design Institute. It publishes long-form essays that trace the shift from screen-based interaction to delegated action \u2014 covering trust architecture, delegation design, agentic commerce, machine customers, agentic AI protocols, and human-agent interaction. Founded by Tony Wood in September 2024, it has published 64 research essays and is the foundational research record of the AXD discipline.
The Observatory covers six topic clusters: (1) Trust and Delegation \u2014 trust architecture, delegation design, trust recovery, temporal trust, and trust debt. (2) Agentic Commerce and Payments \u2014 machine customers, agentic shopping, agent payments, Stripe Five Levels, Mastercard vs Visa, x402, and Verifiable Intent. (3) Protocols and Infrastructure \u2014 MCP, A2A, ACP, Universal Commerce Protocol, and agentic entity resolution. (4) Design Principles \u2014 operational envelopes, interrup
New readers should start with the founding essay 'Beyond the Interface
The Observatory provides the intellectual foundation that the Practice section operationalises. Each essay develops concepts and arguments that inform the 12 design frameworks, the AXD Playbook, and the role-specific How To guides. The essays also respond to real-world developments \u2014 when Stripe publishes x402 or Mastercard launches Agent Pay, the Observatory analyses the implications through the AXD lens, keeping practitioners current.
The most recent additions to the Observatory are the Bridging Essays (Issues 055-062), which connect AXD theory to commercial practice: The Economics of Trust examines why trust architecture is the primary commercial lever. The Post-Purchase Problem addresses returns, subscriptions, and loyalty when the customer is an agent. The Value Chain Redesign maps platform economics and ecosystem control. The Agent Taxonomy classifies commerce agents by delegation pattern. The Merchant's Stack examines OM
AXD Observatory: Essays on Agentic Commerce and AI Design The Observatory is the research record of a discipline being built in real time. It brings together long-form essays that trace the shift from screen-based interaction to delegated action, helping readers understand not just what is changing, but why the change matters structurally. Issue 001 · September 2024 · 12 min read Why the Age of UX is Ending and What Must Come Next I started writing the Observatory in September 2024 with a single conviction: the age of agentic AI would require a design discipline that did not yet exist. The first essay, The early essays laid the theoretical foundations. The middle period turned from theory to architecture. I wrote about trust recovery, trust debt, temporal trust, and autonomous integrity - the dynamics of relationships that accumulate history. I wrote about agent observability, interrupt frequency, and the relational arc - the design of oversight in systems that operate in your absence. I wrote about composable interfaces, A2UI, and the invisible layer - what happens to the interface when the best experience is no experience at all. Then the industry arrived. In 2025, Anthropic published MCP, Google published A2A, and OpenAI published ACP. Stripe published their Five Levels of Agentic Commerce and the x402 protocol. Mastercard launched Agent Pay. Visa launched Intelligent Commerce. I wrote analyses of each, identifying the transaction bias and the trust vacuum that their models leave unaddressed. Don Tapscott's identic AI concept, Know Your Agent, agentic entity resolution, intent engineering, agentic markdown - each essay responded to a real development with an AXD analysis that no one else was providing. The most recent series - the Bridging Essays (Issues 055 to 062) - represent a deliberate turn. The first fifty-four essays built the theoretical architecture: trust, delegation, observability, protocols, readiness. The Bridging Essays connect that ar