Retailers Pilot Agentic AI Shopping Agents at Scale

Per the Payments Optimization Tracker report from PYMNTS Intelligence and Worldpay, 43% of retailers are piloting autonomous AI shopping agents, and 81% of retailers indicate they trust AI's ability to act autonomously when the right guardrails are available. The report also finds 45% of consumers would be comfortable letting AI agents complete purchases, rising to 54% among Gen Z, and 95% of consumers report at least one concern about agentic commerce, with 5% reporting no concerns. PYMNTS reports 50% of U.S. consumers would trust agentic commerce more if fraud protections were in place, and the report projects a $1.7 trillion market opportunity by 2030.

Agentic AI in commerce describes systems that can search, compare, and transact on behalf of users, shifting work from humans to automated agents. Industry-pattern observations: deployments of agentic systems typically raise fraud, explainability, and user-control tradeoffs that require integrated risk controls, logging, and transparent UX flows before consumers delegate purchases at scale.

The PYMNTS findings combine retailer experimentation with persistent consumer concerns, framing adoption as an interplay between technical capability and trust. Industry reporting frames the commercial opportunity as large, but conditional on strong fraud protections and comprehensible decision paths that users can audit.

• •Retail pilot outcomes and how payment providers instrument fraud detection for agentic flows.
• •UX affordances that give consumers clear visibility and reversibility of agent actions.
• •Regulatory and payments-industry guidance on liability and authentication for autonomous transactions.

Editorial analysis: For practitioners, the immediate priorities when building agentic commerce are measurable trust signals, robust fraud telemetry, and UX patterns that let consumers retain control without nullifying the productivity gains agents promise.