Google: AI Checkout Won’t Raise Prices

Google has pushed back against criticism suggesting its upcoming AI-driven checkout tools could be used to raise prices for individual shoppers, insisting that this is not how the system is designed to work.

The debate centres on Google’s plans to expand shopping features within AI Mode and Gemini, including references in its roadmap to “cross-sell and upsell modules”. Critics have interpreted this language as opening the door to personalised price increases based on user data, sometimes described as “surveillance pricing”.

Concerns were first raised by Lindsay Owens, executive director of the US-based consumer economics group Groundwork Collaborative. She argued that “personalised upselling” could involve analysing a user’s chat history and behaviour to charge them more at checkout. The criticism gained wider attention after US Senator Elizabeth Warren echoed the claim, accusing Google of helping retailers push people into spending more money by using their personal data.

Google responded publicly via its official “News from Google” account, firmly rejecting the accusations. The company said claims that its AI checkout tools could be used to increase prices are incorrect and misleading. It stressed that merchants are not allowed to display higher prices on Google than those shown on their own websites.

According to Google, the term “upselling” has been misunderstood. The company says it refers to a standard retail practice of highlighting premium versions or additional options that shoppers may want to consider, rather than increasing the price of the same product. In other words, users may be shown higher-end alternatives, but not charged more for the same item because of who they are.

Google also addressed concerns around its “Direct Offers” pilot, which is part of its wider commerce plans. It said this feature can only work in the customer’s favour, either by offering a lower price or adding benefits such as free delivery. Google stated clearly that Direct Offers cannot be used to increase prices under any circumstances.

Much of the controversy stems from language used in Google’s Universal Commerce Protocol (UCP) roadmap. This document mentions future plans for “native cross-sell and upsell modules” designed to deliver personalised recommendations based on user context. Critics have linked this to fears about individualised pricing, particularly when combined with references to “dynamic pricing” in Google’s technical documentation.

Google has not denied that its AI shopping tools will support features such as real-time stock checks, fast transactions and flexible pricing structures. However, it maintains that these capabilities are aimed at improving the shopping experience and reducing friction, not enabling unfair or hidden price increases. Retailers, Google says, remain responsible for pricing and remain the seller of record.

The issue matters because Google is increasingly positioning AI Mode and Gemini as places where users can move seamlessly from product discovery to purchase. As soon as transactions enter the picture, questions around transparency, pricing rules and consumer protection become far more important.

Google’s long-standing policies in Merchant Center already require price accuracy, preventing situations where shoppers see one price in search results and another at checkout. The company appears to be relying heavily on these existing rules to reassure critics that its AI-powered shopping plans will not undermine trust.

Looking ahead, much will depend on how these tools are implemented in practice. If new policies or technical requirements place extra burdens on retailers, the impact will be felt quickly. Equally, if AI-driven checkout significantly reduces drop-off between browsing and buying, Google is likely to accelerate its rollout and make these features a core part of its AI shopping strategy.