The consumer experiment is becoming routine
Artificial intelligence is no longer a novelty in retail. It is quickly becoming part of everyday shopping behaviour, with Omnisend’s survey indicating that 63% of U.S. consumers now use AI for shopping-related tasks, up from 59% in August 2025. The broader significance is not just higher usage, but growing behavioural normalisation. When four in ten U.S. respondents say they have already made purchases directly inside ChatGPT, AI is no longer operating at the edges of commerce. It is beginning to function as a transactional environment in its own right.
Table of Contents
That shift is reinforced by rising comfort with delegation. Eight in ten U.S. consumers say they would be comfortable letting AI handle transactions for them, suggesting that the market is moving beyond experimentation with recommendations and into the early stages of automated purchasing. The real transition is from AI as a helper to AI as an intermediary, and that has important implications for how retailers compete for visibility, trust and conversion.
Convenience is driving adoption faster than brand loyalty
The strongest use cases remain practical rather than aspirational. U.S. consumers are turning to AI primarily for product research and recommendations, deal-finding, review summaries, gift brainstorming and shopping lists. These are not fringe behaviours. They reflect the most familiar frictions in digital commerce: too many choices, too much repetitive searching and too much cognitive effort. AI is gaining traction because it promises to compress those burdens into a simpler decision path.
The motivations behind this behaviour are equally revealing. Consumers say they use AI to save time, make shopping easier, discover products they would not otherwise find and reduce decision fatigue. Retail discovery is increasingly being valued not for its serendipity, but for its efficiency. That helps explain why 42% of U.S. respondents say ChatGPT already provides better product recommendations than search engines. For many users, AI is beginning to outperform traditional search not by being more comprehensive, but by being more decisive.
Better recommendations come with a growing privacy trade-off
The survey also shows that consumers are willing to hand over meaningful personal data in exchange for better recommendations. Browsing history, purchase history, location data and even email receipts are all information many respondents would allow an AI shopping assistant to access. This signals that shoppers increasingly understand personalisation as a transaction: better relevance in return for deeper visibility into their habits.
But willingness is not the same as trust. The same consumers embracing AI shopping are also acutely aware of its risks. Their biggest concerns centre on how data is collected and used, whether AI might complete purchases without approval, and whether prices or product choices could be optimised for the platform instead of for the shopper. Fears about biased recommendations and sponsored placements further underline that consumers do not yet see AI as a neutral assistant. They see it as potentially useful, but also potentially manipulative.
Retailers will be judged on control, transparency and restraint
That tension defines the commercial challenge ahead. Consumers are open to AI-enabled shopping, but only within boundaries they can understand and control. The survey makes those expectations clear: strong privacy protections matter most, followed by the ability to review and approve decisions before purchase. More accurate personalisation helps, but so does clarity about why products are being recommended and proof that recommendations are not simply paid placements in disguise. Trust will not be built by AI capability alone, but by visible safeguards around that capability.
Retailers should take particular notice of the limits consumers are already drawing. Seven in ten U.S. respondents say they would stop engaging with a retailer that used AI to charge different customers different prices for the same product. That is a sharp warning against using AI merely to optimise extraction. Consumers may accept automation, but they will not easily accept algorithmic unfairness. The fact that only 13% of respondents have used an AI feature built directly into a retailer’s website also suggests that retailers have not yet secured the primary relationship. For now, much of the momentum sits with third-party AI platforms rather than with merchants themselves.
The balance of power in retail discovery is starting to shift
The most important conclusion from the data is that AI is becoming a mainstream layer of commerce before retailers have fully established the rules of engagement. As consumers increasingly rely on AI to research, compare, summarise and even buy, the platforms mediating those choices gain influence over which products are seen, trusted and purchased. That makes AI less a feature of retail and more a new gateway into retail.
For merchants, the strategic question is no longer whether shoppers will use AI. They already are. The question is whether retailers can remain credible and visible within AI-mediated commerce without surrendering customer trust. The winners are likely to be those that treat AI not as a shortcut to conversion, but as an extension of customer experience that must remain transparent, fair and permission-based. As shopping becomes more automated, trust may become the most valuable part of the transaction.
Author:
Jan Bielik
CEO & Founder of Webiano Digital & Marketing Agency
Source: Artificial intelligence becomes mainstream shopping tool
