A quiet shift has taken hold in how American consumers begin a shopping trip. Rather than typing a product name into a search bar or scrolling through an online store, a growing number of shoppers now open a chatbot, describe what they want, and let an AI agent handle the legwork. Traffic referred by these AI systems to U.S. retail websites climbed 393% year over year in the first quarter of 2026, according to Adobe Analytics data cited by industry researchers. That kind of growth, from a small base just a year earlier, signals a genuine change in habit rather than a passing curiosity.
The numbers back that up from several directions. Nearly 60% of consumers say they have already used AI to shop, and 97% of retailers report they have either rolled out AI tools or have a program underway, according to consumer research firm Capital One Shopping. Adobe’s own survey found 38% of consumers have used generative AI tools specifically for online shopping, with planned usage climbing well above 50% heading into next year. The pattern researchers keep describing is the same one: shoppers used to discover a product, research it, then buy it. Now research often comes first, discovery follows, and the purchase happens almost as an afterthought.
That broader trend is exactly what CI&T Inc. (NYSE: CINT), a Brazil founded technology and digital transformation firm that helps large companies build AI driven products, set out to measure in a consumer survey released today. The company, which serves clients across financial services, retail, and consumer goods, publishes regular research on how shopping behavior is changing. Its latest report, titled Retail Tech Report Agentic Commerce Edition, focused on how U.S. consumers are using AI agents and what that means for the businesses trying to sell to them.
The findings suggest AI agents have moved from novelty to mainstream. According to the report, 90% of respondents have either used an AI agent while shopping or are open to trying one, and 74% say they already have. Among the remaining holdouts, 63% said they would be willing to give it a try. Just as notable, 78% of respondents said AI makes them smarter shoppers, and 55% said it is actually leading them to shop more, not less. Consumers are mainly turning to these agents for research, comparing brands, hunting for the lowest price, and locating where a specific item is sold.
Melissa Minkow, CI&T’s Global Director of Retail Strategy and Insights, described the shift as a reordering of the traditional shopping journey, noting that the old sequence of discovering a product, researching it, and buying it has flipped so that research now often happens first. She added that AI agents are not replacing retailer websites so much as becoming the first stop shoppers make before they ever land on one, which puts pressure on retailers to make their products easy for AI systems to find and recommend. Electronics, personal care, and apparel are among the categories seeing the fastest AI adoption, the report found.
For a company like CI&T, whose business depends partly on selling digital transformation and AI integration work to retailers, publishing this kind of research doubles as a way of demonstrating expertise in the exact trend its clients are trying to navigate. The stock has had a rough year regardless, trading around $3.20, well below its 52 week high of $5.88. Whether the agentic commerce narrative changes how investors view the company is a separate question from whether the underlying consumer trend is real, and on that point the data across several independent research firms is largely consistent.
Retailers now face a customer that does not browse the way people used to. It compares prices instantly, reads reviews without getting tired, and reports back to a human who may never visit the retailer’s website directly. That changes what it means to be discoverable, since ranking well with a search engine is no longer enough if an AI agent cannot easily read and recommend a product. The businesses that adapt fastest to being understood by machines, not just people, are likely to capture a disproportionate share of this shift.
