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 |
B2B agentic commerce is the application of autonomous AI agents to business-to-business transactions - procurement, supply chain negotiation, vendor evaluation, contract management, and institutional financial services. Unlike B2C agentic shopping, B2B agentic commerce involves organisational agents operating within complex approval hierarchies, compliance requirements, and multi-stakeholder governance structures.
B2B agentic commerce will scale before B2C because enterprise procurement is already highly systematised with structured decision criteria, established governance frameworks, and high transaction volumes that create strong ROI incentives for automation. B2B environments also have existing trust architecture - approval hierarchies, spending limits, audit trails - that maps directly onto the AXD concepts of delegation design and constraint specification.
Machine customers in financial services are autonomous AI agents that execute trades, manage portfolios, process payments, negotiate credit terms, and conduct institutional banking operations on behalf of organisations. The critical design challenge is consequence asymmetry - higher-consequence financial decisions require tighter constraints, more frequent human checkpoints, and comprehensive audit trails. Agentic KYC processes must verify the entire delegation chain from human principal to auto
B2B trust architecture differs from B2C in four key ways: multi-stakeholder delegation (organisational hierarchies rather than individual authority), compliance-embedded trust (regulatory requirements as structural constraints), comprehensive audit trails (legal requirements beyond transparency), and counter-party agent interaction (both sides may be represented by autonomous agents). B2B trust architecture must accommodate collective governance rather than individual risk tolerance.
The AXD approach to B2B agent design centres on five principles: organisational delegation hierarchies (agents respect existing authority structures), multi-agent coordination (protocols for multiple agents to collaborate), graduated autonomy (expanding agent authority as trust accumulates), regulatory compliance by design (compliance as structural constraint), and human escalation protocols (clearly defined paths for human judgment on complex decisions).
The reasons are structural. Enterprise procurement is already highly systematised - purchase orders, approval workflows, vendor qualification processes, and contract templates create the structured environment that autonomous agents require. Consumer shopping, by contrast, is characterised by subjective preferences, emotional decisions, and context-dependent choices that are far harder to delegate. An agent can evaluate vendor bids against a procurement specification more reliably than it can decide whether a consumer will like a particular shade of blue. B2B transactions also have higher stakes and higher volumes, creating stronger economic incentives for automation. A procurement team processing 10,000 purchase orders per month has a clear ROI case for autonomous agents. A consumer making 10 purchases per month has less incentive to invest in agent infrastructure. The economics of B2B agentic commerce are compelling precisely because the transactions are repetitive, high-volume, and governed by explicit rules. Furthermore, B2B environments already have the The trust architecture required for B2B agentic commerce differs fundamentally from B2C agent delegation. In B2C, a single human delegates to a single agent with personal preferences and individual risk tolerance. In B2B, an In B2B agentic commerce, the delegating entity is not a person but an organisation. This means that delegation authority must flow through organisational hierarchies - a procurement manager may authorise an agent to purchase office supplies up to a certain value, but capital expenditure requires board-level approval. The agent must understand and respect these hierarchical delegation boundaries. This is fundamentally different from B2C, where a single person grants and constrains authority. B2B transactions operate within regulatory frameworks - anti-money laundering (AML), know-your-customer (KYC), sanctions screening, data protection, and industry-specific regulations. Autonomous B2B agent