Recent research indicates that while consumers are increasingly turning to artificial intelligence for holiday gift ideas and product research, turning over the reins of actual purchasing power to an autonomous bot is still a bridge too far for most buyers. According to a comprehensive global survey by Salesforce covering more than 2,500 shoppers across the United States, the United Kingdom, Canada, and Australia, 74 percent of consumers trust recommendations provided by AI agents. Despite this high level of advisory trust, checkout remains an entirely separate challenge for retailers looking to automate the full retail journey.

The Rise of AI Advisory and Brand Discovery

The study reveals that 20 percent of holiday shoppers worldwide are already utilizing AI agents as their primary starting point for research and shopping advice. This shift is driving substantial traffic increases to e-commerce platforms, with referral traffic from AI tools surging by 104 percent year-over-year. Furthermore, Salesforce forecasts that AI-driven product recommendations will influence roughly $75.8 billion in Cyber Week spending alone, marking a 13 percent increase compared to previous metrics.

Beyond general recommendations, AI agents are actively reshaping brand discovery. Approximately 30 percent of surveyed consumers reported that an AI agent introduced them to a brand they had never encountered before. However, this advisory power cuts both ways: about 11 percent of shoppers noted they would actively avoid a brand if an AI agent explicitly advised them to do so. Retailers integrating AI shopping solutions are observing notable benefits, including orders that are eight percent larger and contain 23 percent more items on average, alongside a twofold increase in the likelihood of a shopper completing a purchase.

Why Consumers Draw the Line at Autonomous Checkout

Despite strong enthusiasm for research, price comparisons, and gift brainstorming, consumers remain deeply hesitant to let autonomous agents handle financial transactions. Only 19 percent of personal agent users currently employ logic to allow bots to complete checkouts, and merely 9 percent grant their AI tools unrestricted access to sensitive financial details. To mitigate potential security risks, 36 percent of consumers have established separate credit cards exclusively for their AI tools, while 73 percent express outright discomfort with sharing their standard payment information.

Security concerns are compounded by the low tolerance for errors. One in three consumers will completely abandon an AI agent after encountering just two or three mistakes, and a similar proportion will place direct blame on the technology if it yields a poor recommendation. Nearly half of all surveyed shoppers state they have zero interest in allowing an agent to make financial transactions on their behalf due to fears regarding unwanted charges or the exposure of purchase histories.

The Autonomous Retail Future

Looking ahead, industry analysts emphasize that modern retailers must continue to invest in agentic AI capabilities to remain competitive. As consumer agents grow more sophisticated, merchants need systems capable of detecting buying intent, signals of dissatisfaction, and shifting consumer patterns at machine scale. Although current users primarily treat personal bots as sophisticated advisors rather than autonomous buyers, the overarching trajectory of digital commerce points toward a future where automated transactions become frictionless, highly personalized, and standard practice for shoppers worldwide.