Agent-assisted consumer spending will reach $3-5 trillion by 2030, with B2B spending potentially tripling that figure, according to Mastercard's new Signals report. That scale of opportunity comes with a hard requirement: brands must earn the trust of AI agents that increasingly sit between them and their customers.
The report, "Encoding Trust: The Race for Intent, Consent and Control in the Agentic World," found that while 85% of consumers are open to working with AI agents to find the best product, only one in ten will let an agent complete a purchase fully on its own. The gap defines the next challenge for marketers.
The shrinking human funnel
Google has already launched a universal commerce protocol that lets AI agents transact across retailers. Amazon released tools allowing agents to browse other retailer sites on a customer's behalf. These shifts mean traditional marketing funnels could become a conversation between a brand and a machine, not a brand and a consumer.
Mastercard's research suggests autonomy alone won't drive adoption. "Agentic commerce will properly scale when we can prove that every action is bound by consent, is visible to the user, and reversible if something goes wrong," the report said. "It's about designing confidence into the flow."
AI agents may handle routine purchases within set parameters like a budget, while consumers retain approval over exceptions such as unexpected price changes. That hybrid model keeps humans in the loop but shrinks the moments where traditional marketing messages reach a living buyer.
What persuades an algorithm
Marketers have spent decades refining persuasion tactics for human shoppers. Those same techniques produce inconsistent results with AI agents. Research published by Harvard Business Review found that countdown timers, scarcity messages, price anchoring, and FOMO tactics had little to no influence on AI agents compared to humans.
An OSF survey of 50 e-commerce executives revealed most companies have noticed conversion shifts attributed to AI agents. Yet many still believe the same cues that persuade humans also influence machines. The data suggests otherwise.
Competitive pricing and strong authentic ratings carry more weight with AI agents. Brands also need to recognize that different AI models respond differently to marketing cues, making it necessary to identify which models drive the most traffic and purchases. Understanding the prompts consumers give to AI agents may become a new form of customer research, revealing what shoppers ask AI to prioritize during discovery.
For marketing professionals adapting to this shift, structured training can close the knowledge gap. An AI Learning Path for Marketing Managers addresses how to build strategies that work when the buyer is an algorithm, not a person browsing a landing page.
Building the trust layer
Mastercard's report proposes a trust infrastructure where systems verify who is acting, what the user authorized, and whether the user's rules are being followed, while maintaining a record of the agent's activity. Agentic tokens could provide specific permissions, such as spending limits, rather than giving AI unlimited access to payment credentials.
This infrastructure creates new demands on brands. Machine-readable product information may become as important as ad copy. If an AI agent can't parse a product's specifications, warranty terms, or sustainability claims, that product simply won't surface in agent-driven searches.
Continuous testing and adaptation of product information will likely replace fixed, traditional approaches. Brands that treat product data as a living asset, updated as AI models evolve, stand a better chance of remaining visible in automated ecosystems. Marketers building these capabilities can benefit from dedicated AI for Marketing resources that cover agentic commerce and machine-readable content strategies.
Why this matters for marketing professionals
When an AI agent decides which products to recommend, it doesn't respond to emotional appeals, brand storytelling, or urgency tactics. It responds to structured data, clear permissions, and verifiable trust signals. Marketing teams that continue optimizing for human psychology alone will lose ground to competitors who optimize for machine readability. The immediate action is to audit product information feeds, ensure competitive pricing is accurately reflected in machine-readable formats, and begin tracking which AI models drive traffic to your digital storefronts.
Your membership also unlocks:
