Same Product, 21.6% Higher Price: Inside the Study Exposing Google AI Mode's Shopping Blind Spot

AI search was supposed to eliminate the friction of comparison shopping. Instead, according to one of the largest independent data studies of Google’s AI Mode to date, it may quietly be showing you more expensive versions of the very products you were already looking at.

Over 23 days — August 9 to 31, 2026 — e-commerce analytics platform Productrise tracked more than 2 million product listings across over 100,000 search engine results pages and AI Mode responses in the US and UK. The methodology was deliberately ruthless in its simplicity: the same shopping queries were run through both traditional Google Search and Google AI Mode on the same calendar day, at the same moment, and products were matched across the two surfaces using Google’s own stable product identifiers.

The headline finding: when the exact same product appeared on both surfaces, the lead price shown in AI Mode averaged 21.6% higher than in traditional search.

What the study actually measured

The cleanest comparison in the dataset is the matched-pair analysis. Same product, same query, same day — the only variable is which Google surface you happened to be looking at. For each pair, Productrise took the lead offer on each side (the lowest listing position) and computed the relative difference.

Across those matched pairs, AI Mode was 21.6% more expensive on average. Zoom out to every priced listing on each side — not just the matches — and the gap widens dramatically: the median product in AI Mode was priced at $149 versus $100 in traditional search, roughly 49% higher.

The asymmetry gets sharper when you look at disagreement cases. Matched products showed different prices 38.1% of the time. When they disagreed, AI Mode was the pricier surface 68.4% of the time — about two-thirds — with a median difference of 22.2%. The average difference in those cases was a striking 88.5%, though Productrise notes that figure is skewed by outliers, such as a used listing in traditional search matched against a new one in AI Mode.

When the situation reversed and AI Mode was actually cheaper, the gap was notably smaller: a median of 7.8% and an average of 11.9%. In other words, AI Mode’s pricing penalty when it loses is bigger than its discount when it wins.

A different shelf, not just a different price tag

Two structural findings make the price data more consequential.

First, overlap between the two surfaces is nearly nonexistent. Only 1.28% of products ranking in traditional search also appeared in AI Mode for the same search on the same day. AI Mode showed an average of just 3.9 products per query versus 27.8 in traditional search. This echoes Productrise’s July 2026 study, which found AI Mode returns roughly 95% fewer product listings than traditional search on identical queries.

Second, even when the same product ranks on both sides, AI Mode picks a different lead seller 49.6% of the time. That detail matters more than the price gap itself, as Callum Lockwood, director of organic search at Re:signal, put it: it suggests AI Mode “isn’t just re-ranking the same offer, it’s picking a different one on its own criteria” — feed data, availability, third-party citations, reviews, and press coverage all likely feed into what the model chooses to recommend.

Why this lands harder than a normal ranking quirk

Google has spent the past year pulling AI Mode into the default search experience. A prominent AI Mode button now sits directly on the standard search box. Chrome desktop users see it in the address bar. Search results carry an AI Mode tab that carries the same query straight over. In most cases, the switch happens without any deliberate choice by the shopper — from their point of view, nothing appeared to change.

That makes the first price shown disproportionately important. Clicking a product in either surface opens a knowledge panel listing other sellers and their prices, sometimes cheaper ones — but that takes a deliberate click most shoppers never make. The lead listing acts as an anchor, and per this data, the anchor in AI Mode is systematically set higher.

Independent SEO consultant Brodie Clark noted the irony: traditional Google product listings have historically rewarded the cheapest offer with the best visual treatment, and Google has rolled out features specifically for price-conscious shoppers. “Those same pricing treatments are effectively forgotten in the AI Mode examples,” he observed. Katelyn Geary, senior SEO strategist at Break The Web, framed the consumer stakes bluntly: “We’re trading the visible effort of opening multiple windows for the invisible penalty of paying more without ever seeing the cheaper option that was one tab over.”

The merchant side of the story

Not everyone sees this as pure downside. Jamie D’Alessandro, SEO manager at JAKALA, pointed out that AI Mode appears more likely to link to a brand’s own website over third-party marketplaces — potentially a significant shift in where shopping traffic lands. And for brands that cannot win a race to the bottom on price, a discovery surface that doesn’t gate visibility on lowest-price is arguably the first opening they’ve had in years.

The study’s author, Productrise founder Hugo Huijer, draws the strategic conclusion: the cheapest price is demonstrably less of a ranking factor in AI Mode, which “changes who gets to compete.” Richer product feeds, better product data, reviews, and alignment with shopping conversations can make an AI confident enough to recommend a product that would never have survived the traditional carousel’s price-sorting logic.

Caveats worth keeping in mind

The study comes from a commercial platform that sells tracking tooling for organic product visibility, so it has a natural interest in AI Mode being newsworthy. Google has not, as of this writing, offered a detailed public response to the findings. The SERP-side comparison was limited to popular_products carousel listings rather than full results, and the 88.5% average gap is explicitly outlier-driven. It is also a snapshot of a surface Google is still actively evolving.

But the core matched-pair methodology is sound, the sample is large, and the direction of the effect is consistent across every cut of the data. Coverage by PCMag, Futurism, Search Engine Journal, and MediaPost has pushed the findings into mainstream view this week.

For shoppers, the practical takeaway is simple: the number AI Mode shows you first is not necessarily the best number available. For brands, it’s that “fact-checking AI” in shopping experiences — knowing what the model recommends and at what price — just became a core discipline, not a nice-to-have.