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 |
Autonomous shopping is the practice of AI agents independently discovering, evaluating, comparing, negotiating, and purchasing goods or services on behalf of a human principal - without requiring moment-to-moment human oversight. It exists on a five-level autonomy gradient, from filtered discovery (Level 1) to full autonomy (Level 5), with trust architecture requirements increasing at each level.
Agentic shopping is the broader category - any shopping activity mediated by an AI agent, regardless of autonomy level. Autonomous shopping is a subset of agentic shopping, specifically the levels (3-5) at which the agent exercises independent decision authority without moment-to-moment human oversight. The design requirements differ: assisted agentic shopping needs good recommendations, while autonomous shopping needs trust architecture, delegation design, and recovery mechanisms.
The autonomy gradient describes five levels of shopping agency: Level 1 (Filtered Discovery - agent filters, human decides), Level 2 (Recommended Selection - agent recommends, human decides), Level 3 (Constrained Autonomy - agent decides within defined boundaries), Level 4 (Supervised Autonomy - agent operates independently with periodic review), and Level 5 (Full Autonomy - agent anticipates needs and completes purchases without human involvement).
Autonomous shopping requires four types of trust: delegation trust (the human trusts the agent enough to hand over purchasing authority), execution trust (the agent consistently makes good purchasing decisions), recovery trust (the agent handles mistakes through proactive disclosure and correction), and evolving trust (the agent progresses up the autonomy gradient as it demonstrates competence). Trust architecture is the structural foundation that makes autonomous shopping possible.
A machine customer is an autonomous shopping agent operating at Levels 4-5 of the autonomy gradient - the point at which the agent is effectively a customer in its own right, interacting with merchants and marketplaces without human involvement. Machine customers require merchant-side design changes including agent-readable product information, machine-negotiable pricing, and programmatic purchasing interfaces.
The critical distinction is the locus of decision authority. In assisted shopping, the human decides and the agent helps. In autonomous shopping, the agent decides and the human has delegated. This shift from human decision to agent decision is the defining characteristic of autonomous shopping - and it is the reason that autonomous shopping requires a fundamentally different design approach from traditional e-commerce or AI-assisted shopping. Autonomous shopping is not a binary state - it is a gradient. The The design of the autonomy gradient is the central challenge of autonomous shopping. Too little autonomy, and the agent provides insufficient value - the human might as well shop themselves. Too much autonomy, and the human loses control - the agent may make purchases the human would not have chosen. The optimal position on the gradient depends on the domain (groceries vs luxury goods), the consequence level (low-cost vs high-cost), and the accumulated trust between the human and the agent. The AXD Institute defines five levels of shopping autonomy, each representing a distinct design challenge and trust architecture requirement: The agent filters and organises information based on the human's stated preferences, but the human makes all decisions. The agent surfaces relevant products, removes irrelevant options, and organises results by the human's criteria. Trust requirement: minimal - the agent's errors are easily corrected by the human. Design focus: preference learning and filter accuracy. The agent recommends specific products or services, providing reasoning for each recommendation. The human reviews recommendations and makes the final decision. Trust requirement: moderate - the human must trust the agent's judgment enough to consider its recommendations seriously. Design focus: recommendation transparency and reasoning legibility. The agent selects and purchases within defined constraints - budget limits, brand preferences, quality thresholds, delivery re