trending
The first trend in Global Payments’ 2026 Commerce and Payment Trends Report marks a decisive shift in how transactions are initiated and executed: AI is evolving from a passive recommendation engine into an active purchasing agent.
AI as buyer
What began with chatbots handling customer queries has matured into “agentic commerce”, where AI systems can research, decide and complete purchases on behalf of consumers, often without repeated human intervention.
This transition alters not just the shopping journey, but the very locus of decision-making. As the report notes, more than 85% of surveyed businesses are already familiar with the concept, signalling that agentic commerce is no longer theoretical but an emerging operational reality.
Why Agentic Commerce Matters to Payments
The commercial implications are substantial. Cart abandonment remains one of ecommerce’s structural inefficiencies, with roughly two-thirds of baskets never converting. Agentic commerce promises to compress that leakage by automating intent into execution.
By linking discovery, preference-setting and payment into a single flow, AI agents can schedule repeat purchases, manage subscriptions and dynamically adjust orders based on inventory or pricing signals.
For payments providers, this reframes the transaction itself.
The question is no longer how fast or cheap a payment clears, but how seamlessly payment can be embedded into an autonomous decision loop. As the report highlights, this is where orchestration, interoperability and secure tokenisation become strategic assets rather than backend utilities.
Merchants, Data and the New Information Asymmetry
Agentic commerce also changes what merchants know about customers. Traditionally, sellers inferred intent from clicks and cookies.
AI agents, by contrast, can explicitly share structured insights — budget constraints, brand preferences, purchasing history — if consumers opt in.
This creates a richer, permissioned data exchange that can improve conversion, loyalty and personalisation, particularly for small and medium-sized businesses seeking new digital distribution channels.
Yet this advantage cuts both ways. Merchants must now optimise not only for human users but for machine readers. Websites, catalogues and APIs increasingly need to be “AI-readable”, echoing the mobile optimisation wave of the mid-2000s.
Trust, Control and the Security Question
If agentic commerce is to scale, trust will be its binding constraint. The report is clear that security concerns dominate business hesitations, from fraud prevention to dispute resolution.
Network tokenisation, biometric authentication and explicit consumer instructions embedded in authorisation flows are positioned as critical safeguards, allowing AI agents to act while remaining accountable to human intent.
Ultimately, agentic commerce will not succeed by removing humans from the loop entirely, but by redefining their role.
Consumers may delegate execution, but they will still demand control, reversibility and recourse. Payments, in this future, become less about moving money and more about encoding trust into autonomous systems.
Comments