The checkout line of the future has no line at all. Nearly half of consumers now use artificial intelligence tools during their shopping journeys, and the infrastructure powering those interactions — autonomous AI agents that can discover, compare, negotiate, and even purchase products on a shopper\u2019s behalf — is advancing faster than most retailers anticipated. Welcome to the age of agentic commerce, where the storefront is a conversation and the sales associate is a large language model.
The Numbers Behind the Shift
The scale of the opportunity is staggering. McKinsey estimates that agentic commerce could orchestrate as much as one trillion dollars in US retail revenue by 2030, and between three trillion and five trillion dollars globally. Morgan Stanley puts a more conservative frame on the near term, forecasting that AI-driven shopping agents could account for 190 billion to 385 billion dollars in US e-commerce spending by the end of the decade, representing 10 to 20 percent of total online sales. Bain projects that 15 to 25 percent of all e-commerce will flow through agentic channels by 2030.
The market for agentic AI in retail and e-commerce already stands at 60.43 billion dollars in 2026, according to Mordor Intelligence, and is growing at a compound annual rate of 29.29 percent, on pace to reach 218.37 billion dollars by 2031.
Consumer behavior is shifting in tandem. A Kearney study found that 73 percent of shoppers are already familiar with AI tools, and 60 percent expect to use AI agents within the next 12 months. Adobe Analytics data shows that AI-driven traffic to US retail sites grew 393 percent year over year in the first quarter of 2026, with AI-referred shoppers converting 42 percent better than those arriving through traditional channels.
The Discovery-to-Purchase Gap
Yet the revolution comes with a caveat. While consumers are enthusiastic about using AI for product discovery — 45 percent use AI assistants for product ideas, 37 percent for summarizing reviews, and 32 percent for comparing prices — trust drops sharply when it comes to handing over the credit card. Only 14 percent of US consumers currently trust AI to place orders on their behalf, according to YouGov data, with trust highest among Gen Z at 29 percent and millennials at 30 percent.
"When the journey gets complicated, shoppers don\u2019t just get frustrated — they bounce," said Carrie Tharp, Vice President of Global Solutions and Industries at Google Cloud, speaking at the National Retail Federation\u2019s Big Show earlier this year. "Seamless experiences are now table stakes for a healthy business."
That gap between discovery enthusiasm and transaction trust is partly a function of immature infrastructure. Checkout flows, payment authentication, and identity verification systems were designed for human users clicking through web pages, not for autonomous agents making API calls. Walmart learned this the hard way: in-chat purchases through ChatGPT initially saw conversion rates three times lower than when users were redirected to Walmart\u2019s own website.
The Data Quality Imperative
For retailers, the most urgent implication may be invisible to shoppers entirely. AI agents are far less forgiving than human browsers. Where a customer might overlook a missing product dimension or an inconsistent description, an agent will simply skip the listing. Research from Mirakl shows that 42 percent of customers already abandon purchases due to insufficient product information — and agents automate that abandonment at machine speed.
"Near-term advantage will likely go to merchants whose catalogs are easiest for AI to interpret in natural language," said Jonathan Arena, co-founder of New Generation, a technology company that recently launched Kepler, a platform designed to make retail websites fully AI-ready.
Pages with structured data are cited 3.1 times more frequently in Google AI Overviews, and 71 percent of pages cited by ChatGPT include structured data markup. Yet 45 percent of retailers report recurring data quality issues that affect business decisions. The message is clear: if your product catalog is not machine-readable, your products may never enter the consideration set.
Security and the Trust Equation
The rise of autonomous shopping agents also introduces new vectors for fraud. Accenture reports that 78 percent of financial institutions expect fraud to increase significantly due to agentic commerce. Darwinium\u2019s latest research found that 97 percent of organizations have experienced a rise in AI-facilitated attacks over the past year, suffering average annual direct losses of 4.5 million dollars. Visa has documented a 25 percent spike in malicious bot-initiated transactions over a recent six-month period, with a 40 percent surge in the United States alone.
In response, the industry is developing what McKinsey calls "Know Your Agent" protocols — a framework analogous to Know Your Customer requirements in banking, focused on verifying not just the end user but the AI agent acting on their behalf, including its identity, permissions, and behavioral patterns.
What Comes Next
The trajectory is unmistakable. Gartner predicts that 40 percent of enterprise applications will embed AI agents by the end of 2026, up from less than five percent in 2025. Protocol standards like the Agentic Commerce Protocol are maturing to support multi-item carts, subscription management, and even agent-to-agent B2B procurement. Close to 60 percent of small businesses now use AI, double the share from 2023, according to the US Chamber of Commerce.
For retailers large and small, the strategic calculus has changed. It is no longer sufficient to optimize for human eyes scanning a search results page. The next generation of customers may never visit your website at all — their AI agent will do it for them, and it will make its judgment in milliseconds based on the quality of your data, the responsiveness of your APIs, and whether your systems can complete a transaction without friction.
The brands that win will not necessarily be the biggest or the best known. They will be the ones whose digital infrastructure speaks the language that agents understand.
“Near-term advantage will go to merchants whose catalogs are easiest for AI to interpret.”— Jonathan Arena, Co-founder, New Generation