AI Pricing Isn’t the Enemy Consumers Think It Is

Regulators, retailers and consumer watchdogs have raised alarms over artificial intelligence-powered pricing systems, warning that the algorithms work against buyers.

New research from the University of Pennsylvania and Rutgers Business School suggested the picture is more complicated. In many competitive markets, dynamic pricing moves prices lower, not higher.

The Finding That Cuts Against the Consensus

Prices on online marketplaces are often set by algorithms that adjust based on what rivals charge and how much sellers bid for visibility in search results. Their growing use has drawn warnings from regulators who worry the systems learn to move in lockstep, pushing prices above competitive levels. In March, the House Committee on Oversight and Government Reform opened an inquiry into AI-driven pricing across consumer-facing industries.

Ron Berman of the University of Pennsylvania and Hangcheng Zhao of Rutgers Business School found that picture incomplete. Drawing on data from nearly 2,400 product searches conducted repeatedly on Amazon, the study found that prices fall in categories where algorithmic pricing is widespread and shoppers consider fewer options before buying. The dataset covered more than 2 million products and over 1 million daily price observations.

“A lot of research assumes this is bad for consumers, that algorithms collude on prices and weaken competition,” Berman said, per a Tuesday (May 26) Knowledge at Wharton report.

The Ad Auction at the Center of It

The mechanism is not price competition itself but the advertising layer beneath it, according to the report. Sellers bid for prominent placement in search results. Those bids feed directly into margins and, by extension, into the prices shoppers see. Berman and Zhao found that instead of driving prices up, algorithms end up moving together to push advertising bids lower, cutting merchants’ costs and feeding through into lower prices for buyers.

Advertisement: Scroll to Continue

That dynamic traces back to a structural quirk in how online ads are sold. In auction systems, sellers often overpay for prominent placement, a dynamic known as the “winner’s curse,” the report said. The algorithms learn to avoid that.

The downstream effects also reach the platform. Lower prices trigger sales, raising platform commission revenues enough to more than offset the drop in advertising income, according to the report. Platforms collect a percentage cut on each completed transaction, so volume compensates for thinner ad margins.

Where Shopper Behavior Changes the Math

The pro-consumer outcome is not guaranteed. It depends on how people actually shop.

“When buyers compare fewer products, algorithms push prices below where they would otherwise be,” the report said. “When they compare more, pricing climbs above that level.”

The distinction maps onto product categories. Shoppers typically only scan 15 products, and fewer than 25% reach items in the middle of the page, according to the report.

For commodity purchases like toothpaste or paper towels, where buyers grab the first credible result, ad-bid compression dominates. For electronics, where shoppers search and compare across many listings, algorithms can still push prices higher, the report said.

That split creates a challenge for regulators trying to write rules that hold across categories. Algorithmic coordination raises a concern distinct from traditional price-fixing, according to the report.

“Marketplaces are under heavy scrutiny,” Berman said, per the report. “If they offer these algorithms, regulators may say you are helping sellers collude without them talking to each other.”

For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.