Trademarks

The Knowledge Line: Tiffany v. eBay and the Limits of Marketplace Liability

The Second Circuit held that an online marketplace is contributorily liable for counterfeit sales only when it knows of specific infringing listings—not because counterfeiting is rampant on its platform generally.

February 17, 2025
Person shopping on an online marketplace using a laptop and smartphone
The ruling placed the burden of policing individual listings on the brand, not the platform. Shutterstock
Educational content, not legal advice. This article explains general legal concepts. It does not create an attorney–client relationship. For your specific situation, consult a licensed attorney.

When a luxury house sues the platform instead of the seller, it is asking a court to redraw the boundary of secondary liability. In Tiffany (NJ) Inc. v. eBay Inc., No. 08-3947-cv (2d Cir.), decided April 1, 2010, the U.S. Court of Appeals for the Second Circuit refused to move that boundary in Tiffany’s favor. Affirming a bench-trial judgment by Judge Richard J. Sullivan of the Southern District of New York (entered July 14, 2008 after a week-long trial), the court held that eBay was not contributorily liable for the sale of counterfeit Tiffany silver jewelry on its site, even though the parties agreed that a large volume of counterfeits had in fact changed hands there. The Supreme Court denied certiorari on November 29, 2010, leaving the decision as the foundational American authority on a marketplace’s duty to police counterfeiting.

At a glance

  • Case: Tiffany (NJ) Inc. v. eBay Inc., No. 08-3947-cv (2d Cir.), aff’g 576 F. Supp. 2d 463 (S.D.N.Y. 2008).
  • Decided: April 1, 2010 (Second Circuit); cert. denied Nov. 29, 2010.
  • Court below: S.D.N.Y., Judge Richard J. Sullivan, following a bench trial.
  • Core holding: Contributory trademark infringement requires that the marketplace know or have reason to know of specific instances of infringement; generalized knowledge that counterfeiting occurs on the platform is not enough.
  • Willful blindness: A service provider may not bury its head in the sand, but eBay’s anti-counterfeiting investments (the VeRO program and a “fraud engine”) defeated any claim that it deliberately ignored infringement it had reason to suspect.

The doctrinal frame: Inwood meets the internet

Contributory trademark infringement is a judge-made doctrine, anchored in the Supreme Court’s decision in Inwood Laboratories, Inc. v. Ives Laboratories, Inc., 456 U.S. 844 (1982). Under Inwood, a defendant is liable if it (1) intentionally induces another to infringe, or (2) continues to supply its product or service to one whom it “knows or has reason to know is engaging in trademark infringement.” Tiffany did not seriously press inducement; the fight was over the second Inwood prong as applied to a service rather than a product.

The translation from product to platform is not mechanical. Inwood involved a manufacturer supplying mislabeled pharmaceuticals to pharmacists; later cases, notably Hard Rock Cafe Licensing Corp. v. Concession Services, Inc., 955 F.2d 1143 (7th Cir. 1992), extended the analysis to flea-market operators who provide a venue rather than a good. eBay sat one step further out: it supplied a wholly online marketplace, listing-management tools, and payment facilitation, while exercising some control over what appeared on its site. The Second Circuit accepted that Inwood governs and that eBay exercised sufficient control over the “instrumentality” to be a candidate for liability. The decisive question was the content of the knowledge requirement.

Specific knowledge, not general awareness

Tiffany’s theory was essentially statistical. Its own “Buying Programs”—test purchases of “Tiffany” silver from eBay—suggested that a large share of the listed items were counterfeit, and Tiffany had sent eBay tens of thousands of takedown notices through eBay’s Notice of Claimed Infringement procedure. From this, Tiffany argued, eBay possessed generalized knowledge that counterfeiting was pervasive and therefore had a duty either to police the listings preemptively or to stop offering its services for “Tiffany” sales altogether.

The court rejected the equation of generalized knowledge with Inwood knowledge. It held that “for contributory trademark infringement liability to lie, a service provider must have more than a general knowledge or reason to know that its service is being used to sell counterfeit goods. Some contemporary knowledge of which particular listings are infringing or will infringe in the future is necessary.” Critically, the court found that when eBay received specific notice—through a takedown demand identifying a particular listing—it acted, generally removing the listing within hours. Because eBay responded to the particularized notices it received, the residual counterfeiting reflected sales eBay did not specifically know about, and the law placed no duty on eBay to root those out on Tiffany’s behalf.

This allocation is the heart of the opinion. The burden of identifying particular counterfeit listings rests with the markholder, who is best positioned to distinguish a genuine secondhand Tiffany piece from a fake. The platform’s duty is to act on the specific knowledge it acquires.

Willful blindness as the safety valve

The court did not give marketplaces a blank check. Drawing on Hard Rock Cafe, it confirmed that “a service provider is not … permitted willful blindness. When it has reason to suspect that users of its service are infringing a protected mark, it may not shield itself from learning of the particular infringing transactions by looking the other way.” Willful blindness thus functions as the doctrine’s safety valve: a platform cannot defeat the specific-knowledge requirement by deliberately constructing ignorance.

But on the trial record, eBay was not willfully blind. It had spent tens of millions of dollars on anti-fraud measures, operated the Verified Rights Owner (VeRO) program through which rightsholders could report infringing listings, suspended repeat offenders, and deployed a “fraud engine” of automated filters to flag suspect activity. Those efforts showed an entity trying to learn about and suppress infringement, not one averting its eyes. The willful-blindness theory therefore failed on the facts, and the specific-knowledge rule controlled.

Open questions

  • How much generality is too much? The court drew the line at “particular listings,” but later platforms aggregate data that can make patterns—seller histories, pricing anomalies, image reuse—nearly as probative as a named listing. At what point does sophisticated pattern-detection capability convert “general” awareness into actionable specific knowledge?
  • Does Inwood even fit pure intermediaries? The opinion assumed Inwood applies to online services, but it left unresolved how much “direct control and monitoring” of the instrumentality is required—an issue the Ninth Circuit confronted directly in the web-host context.
  • What about willful blindness in the algorithmic age? If a platform’s own analytics flag a seller as high-risk, does ignoring that signal become willful blindness even absent a brand’s takedown notice?
  • Does the rule survive a different statutory hook? Tiffany is a common-law contributory-liability case. Congress could, in principle, impose a more demanding notice-and-takedown or proactive-screening regime by statute, as the EU has begun to do.

Implications

  • The brand polices; the platform reacts. Tiffany assigns the front-line monitoring burden to rightsholders, who must identify specific listings; platforms owe a duty to act on what they are told, not to hunt independently.
  • Robust takedown infrastructure is a legal asset. eBay’s VeRO program and fraud engine were not just good citizenship—they were the evidentiary record that defeated willful blindness. Modern marketplaces should document their anti-counterfeiting investment.
  • Notices should be specific and actionable. A markholder building a contributory case should send particularized notices and preserve evidence that the platform failed to act on them; bulk statistical assertions of “pervasive” counterfeiting are not enough.
  • The decision shaped global platform norms. Tiffany contrasts with stricter European approaches (e.g., L’Oréal v. eBay in the CJEU), making forum and governing law pivotal for cross-border anti-counterfeiting strategy.

Frequently asked questions

Did the court find that no counterfeit Tiffany goods were sold on eBay? No. The parties agreed that a substantial number of counterfeit “Tiffany” items were sold through eBay. The court held only that eBay’s general awareness of that fact, without specific knowledge of which listings were infringing, did not create contributory liability.

What is the difference between general and specific knowledge here? General knowledge is awareness that counterfeiting occurs somewhere on the platform. Specific (or “contemporary”) knowledge is awareness that a particular listing or seller is offering counterfeit goods. Under Tiffany, only the latter triggers a duty to act.

Can a marketplace escape liability simply by not investigating? No. The willful-blindness doctrine prevents a platform from deliberately avoiding knowledge it has reason to suspect. A marketplace that ignores red flags it actually perceives can be charged with the knowledge it chose not to confirm.

Authorities and sources