万事达卡：2026 AI 智能体与 B2B 支付未来发展白皮书（英文版）（14页）.pdf

1、Agentic AI and the future of B2B paymentsWHITE PAPERJUNE 20262Contents3Executive summary4Reframing B2B payments workflows in an agentic world7What agentic AI actually does in B2B payments9Impact by stakeholder11Supplier enablement:A critical agentic use case12Guardrails,governance,and human oversigh

2、t13ConclusionAgentic AI adds a groundbreaking new layer of delegated automation to B2B payments and acceptance.3Agentic AI adds a groundbreaking new layer of delegated automation to B2B payments and acceptance:agents capable of executing tasks on the customers behalf within carefully defined policie

3、s,approvals and audit trails.While the opportunity is undoubtedly meaningful,the shape adoption takes will be defined by model capability,along with governance,liability management,ecosystem integration and,crucially,the reassurance that agents will operate within authorized boundaries.Value will li

4、kely concentrate first in high-friction workflows,where organizations can restrict what an agent is allowed to do and measure outcomes(e.g.,invoice coding,policy checks,reconciliations,routine purchasing,supplier outreach).Agents are inherently probabilistic,and where customers require highly predic

5、table,low-variability execution,the pattern is usually hybrid:deterministic workflow automation enforces controls,while agents handle unstructured inputs,exception resolution and cross-system coordination.Where the stakes around decisions are higher(e.g.,strategic sourcing,contract commitments,high-

6、value exceptions),the accountability must remain with humans.Adoption will vary across micro and small businesses,mid-market and large enterprise segments because systems of record,integration maturity and governance capacity differ.Segment-specific deployment models will likely be required for scal