Shortly after Google announced its new Universal Commerce Protocol for AI-powered shopping agents, a consumer economics watchdog sounded the alarm. In a now viral post on X viewed nearly 400,000 times, Lindsay Owens wrote that this was big and bad news for consumers. She stated that Google’s plan to integrate shopping into its AI offerings includes ‘personalized upselling,’ which she interprets as analyzing user chat data to overcharge them.
Owens is the executive director of the consumer economics think tank Groundwork Collaborative. Her concern stems from reviewing Google’s roadmap and its detailed specification documents. The roadmap includes a feature supporting “upselling,” which could help merchants promote more expensive items to AI shopping agents. She also called out Google’s plans to adjust prices for programs like new-member discounts or loyalty-based pricing, a point referenced by Google CEO Sundar Pichai when he announced the new protocol.
After inquiries were made about Owens’ allegations, Google publicly responded and spoke directly to reject the validity of her concerns. Google stated that claims around pricing are inaccurate, strictly prohibiting merchants from showing prices on Google that are higher than on their own sites. The company clarified that “upselling” is a standard retail practice for showing premium product options, with the choice always remaining with the user. Google also explained that a pilot program called “Direct Offers” is for lower-priced deals or added services like free shipping and cannot be used to raise prices.
In a separate conversation, a Google spokesperson said the company’s Business Agent does not have functionality to change a retailer’s pricing based on individual data.
Owens also pointed to Google’s technical documents on handling a shopper’s identity, which state that scope complexity should be hidden in the user consent screen. The Google spokesperson told reporters this is not about hiding user agreements but about consolidating actions for user convenience instead of requiring agreement for each one separately.
Even if concerns about this particular protocol are unfounded, as Google asserts, Owens’ general premise is still worth consideration. She warns that shopping agents built by big tech companies could eventually allow merchants to customize pricing based on an analysis of a user’s AI chats and shopping patterns, a practice she calls “surveillance pricing,” instead of charging everyone the same price.
Although Google says its agents cannot do such a thing now, it remains true that Google is fundamentally an advertising company serving brands and merchants. Last year, a federal court ordered Google to change several search business practices after ruling the company engaged in anticompetitive behavior.
While many are excited about a future where AI agents handle tasks, the potential for abuse is clear. The problem is that the big tech companies best positioned to build these shopping tools also have mixed incentives, as their business relies on serving sellers and harvesting consumer data.
This situation means AI-powered shopping could be a significant opportunity for startups building independent technology. Early entrants in this space include startups like Dupe, which uses natural language queries to help people find affordable furniture, and Beni, which uses image and text for thrifting fashion. Until such independent options mature, the old adage probably holds true: buyer beware.