Amazon's Buy Box Steered Shoppers to Costlier Items, a Lawsuit Said. A Judge Dismissed It Anyway.
What happened
A February 2024 lawsuit accused Amazon of using its Buy Box algorithm to route shoppers toward higher-priced listings rather than the lowest available prices, in direct contradiction of what the feature claims to do. Filed in the name of two US-based Amazon customers in the case Taylor et al v. Amazon.com, the complaint alleged that the company violated US consumer protection law by designing the algorithm to serve its own revenue interests ahead of the interests of the shoppers relying on it.
The Buy Box is the mechanism Amazon uses to select which seller's listing gets attached to the prominent "Buy Now" and "Add to Cart" buttons on a product page. Most shoppers never look past it. The complaint argued that the algorithm frequently surfaces items from sellers enrolled in Amazon's Fulfillment By Amazon program, which pay the company fees for inventory storage, packing, and shipping, even when other sellers offer identical products at lower prices with comparable or faster delivery. The result, the plaintiffs claimed, was a recommendation that looked neutral but was not.
The lawsuit drew directly on a concurrent antitrust action against Amazon brought by the Federal Trade Commission and 17 states, which documented that shoppers use Amazon's default selections nearly 98 percent of the time, with many incorrectly assuming the choice reflected the best available price. That figure is load-bearing. A feature that captures 98 percent of purchase decisions and steers even a fraction of them toward higher-priced listings transfers substantial money from buyers to sellers and to Amazon's fee revenue without the buyer ever knowing it happened.
In July 2024, US District Judge Marsha Pechman dismissed the case. Her ruling focused on standing: the plaintiffs had not sufficiently demonstrated how they personally were harmed by the algorithm's selections. The dismissal did not settle the underlying question of whether the Buy Box favors FBA sellers at shoppers' expense. It resolved only whether these particular plaintiffs made a legally sufficient showing of individual harm, which is a narrower bar than it might appear and one that algorithmic systems are structurally well-positioned to avoid.
The outcome points to a core problem in algorithmic accountability. When a recommender system's decision logic is invisible to the people it affects, proving harm becomes nearly impossible, because the shopper who clicked "Buy Now" has no way of knowing what the algorithm chose not to show them. That gap between what a system surfaced and what it had available is exactly the kind of thing a provable record of what a system did would capture. Without one, the algorithm's choices remain beyond scrutiny until a court decides otherwise, and courts have now confirmed that the bar for that is high.
Reported impact
- Affected parties
- Not publicly disclosed
- Harm type
- Not publicly disclosed
- Scale
- Not publicly disclosed
- Financial impact
- Not publicly disclosed
- Regulatory action
- Not publicly disclosed
Classification
Relevant governance controls
Governance control mapping is not available for this record.
- No controls mapped
Not publicly disclosed
Control mapping is analytical. It does not state that any control would have prevented the incident.
Sources and evidence
This record was researched and written by the Index. The event is also catalogued in the following database, which is listed for cross-reference.