Patent infringement is making, using, offering to sell, selling, or importing a patented invention inside the United States without the patent owner’s authority. That definition comes from 35 U.S.C. § 271(a), and it turns on a comparison between the accused thing and the patent’s claims, not between two products. Intent plays no part in it: a company that independently invented the same thing, never read the patent, and never heard of the owner still infringes.
Infringement is also purely a private-law wrong. There is no criminal patent infringement in the United States, no police to call, and no government enforcement. A patent is a right to exclude, and it means nothing until the owner asserts it in court or before the International Trade Commission.
A patent’s claims are the numbered sentences at the end of the document. They define the invention’s boundary in words the way a deed defines land by its edges. The drawings, the abstract, the marketing, and the product the patentee actually sells are not the patent. Only the claims are.
Every infringement analysis therefore runs in two steps. First the court construes the claims, deciding what the disputed words mean as a matter of law. Markman v. Westview Instruments (1996) held that construction is the judge’s job, not the jury’s, which is why the “Markman hearing” is often the moment a patent case is effectively decided. Second, the construed claim is compared to the accused product or process, which is a question of fact.
The comparison is unforgiving in one specific direction. An accused product infringes only if it contains every single element of at least one claim. Miss one limitation and there is no infringement, however similar everything else looks. This is the all-elements rule.
The corollary surprises people: adding features does not help. A claim to a chair with a seat, a back, and four legs is infringed by a chair with a seat, a back, four legs, a cupholder, and a massage motor. More is not different. Less is different.
This is also why claim breadth cuts both ways during prosecution. Every word an applicant adds to a claim to get around prior art is another element an infringer must be shown to have.
Literal infringement means the accused element falls squarely within the claim language as construed. If the claim says “a pH of approximately 6.0 to 9.0,” a process at pH 7 literally infringes.
Because a competitor could otherwise escape by trivial substitution, courts also recognize the doctrine of equivalents: an element that is not literally within the claim still counts if it differs only insubstantially. The classic test, from Graver Tank v. Linde Air Products (1950), asks whether the accused element performs substantially the same function in substantially the same way to achieve substantially the same result.
Warner-Jenkinson Co. v. Hilton Davis Chemical Co., 520 U.S. 17 (1997), preserved the doctrine but disciplined it. Equivalence is assessed element by element, not invention as a whole, so it can never be used to erase a limitation from a claim. That case involved an ultrafiltration process claimed at pH 6.0 to 9.0 and an accused process run at pH 5.0.
The doctrine has a major brake: prosecution history estoppel. If an applicant narrowed a claim during examination for reasons related to patentability, the surrendered territory is generally off limits later. Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722 (2002), rejected an absolute bar and replaced it with a presumption of surrender that the patentee can rebut, for example by showing the equivalent was unforeseeable at the time of the amendment. The practical lesson is that everything said to the examiner becomes public record and can be used to shrink the patent years later.
Someone can be liable without touching the invention, but only if a direct infringer exists somewhere and the defendant knew what was going on.
Induced infringement under § 271(b) covers actively encouraging another’s infringement: instructions, manuals, configuration advice, marketing a product for an infringing use. Global-Tech Appliances v. SEB S.A., 563 U.S. 754 (2011), held that inducement requires knowledge that the induced acts constitute infringement, and that willful blindness satisfies it. The defendant there had a lawyer clear its deep fryer without mentioning it had copied a competitor’s product bought overseas. Commil USA v. Cisco Systems, 575 U.S. 632 (2015), then held that a good-faith belief the patent is invalid is no defense to inducement, because validity and infringement are separate questions.
Contributory infringement under § 271(c) covers selling a component of a patented machine, or material for practicing a patented process, that is a material part of the invention, known to be especially made for infringing use, and not a staple article of commerce suitable for substantial non-infringing use. That last clause is the whole fight. If the component has a real lawful use, there is no contributory liability.
Method claims created a loophole. If a claim requires four steps and one company performs three while its customers perform the fourth, nobody has performed them all, so no one directly infringes. Limelight Networks v. Akamai Technologies, 572 U.S. 915 (2014), confirmed that there is no inducement without a single act of direct infringement.
On remand, the Federal Circuit sitting en banc in Akamai Technologies v. Limelight Networks, 797 F.3d 1020 (Fed. Cir. 2015), closed most of the gap by broadening attribution. One party is charged with another’s performance where it directs or controls that performance, including by conditioning a benefit on performing the step and setting the manner of performance, or where the parties form a joint enterprise. Limelight told customers to tag content and step through its process, which was enough.
Two extraterritorial provisions matter. Section 271(f) reaches supplying components from the United States for assembly abroad, and WesternGeco v. ION Geophysical (2018) allowed recovery of lost foreign profits caused by that domestic act. Section 271(g) reaches importing a product made abroad by a process patented here, which is how U.S. process patents reach offshore manufacturing.
What is patent infringement? Patent infringement is making, using, offering to sell, selling, or importing a patented invention within the United States without the patent owner’s authority, under 35 U.S.C. § 271(a). It occurs only when the accused product or process contains every element of at least one claim, either literally or under the doctrine of equivalents. Intent is irrelevant to direct infringement, so an independent inventor who never heard of the patent still infringes.
What is the all-elements rule? The all-elements rule says an accused product infringes only if it contains each and every limitation of a claim. Missing one element defeats infringement no matter how closely the rest matches, and adding extra features does not avoid infringement. The rule applies limitation by limitation under the doctrine of equivalents too, so equivalence cannot be used to read an element out of a claim entirely.
What is the doctrine of equivalents? The doctrine of equivalents allows infringement where an accused element is not literally within a claim but differs only insubstantially, often tested by whether it performs substantially the same function in substantially the same way to reach substantially the same result. Warner-Jenkinson v. Hilton Davis (1997) confirmed the doctrine while requiring it be applied element by element. Prosecution history estoppel limits it where claims were narrowed during examination.
What is the difference between direct and indirect infringement? Direct infringement under § 271(a) is committed by whoever performs the claimed acts, regardless of knowledge. Indirect infringement requires someone else’s direct infringement plus knowledge. Inducement under § 271(b) covers actively encouraging infringement, and Global-Tech v. SEB (2011) requires knowledge that the induced acts infringe, satisfied by willful blindness. Contributory infringement under § 271(c) covers selling a component with no substantial non-infringing use.
Going further: What to do when someone is infringing your patent .
This page is general legal information, not legal advice, and it does not create an attorney-client relationship.
The Supreme Court held that replacing the worn fabric of a patented convertible top is permissible repair, not infringing reconstruction of the combination.
The Supreme Court's foundational statement of the doctrine of equivalents: a device that works the same way for the same result can still infringe.
A unanimous Supreme Court saved the doctrine of equivalents but disciplined it, requiring an element-by-element test and presuming estoppel when a claim amendment goes unexplained.
The Federal Circuit erased a $106M verdict, holding that canceling a broader claim during prosecution surrenders that subject matter for the doctrine of equivalents.
The Supreme Court held that a patent licensee paying royalties under protest need not breach its license to bring a declaratory-judgment action challenging the patent.
The Supreme Court held that an authorized sale (anywhere in the world, and despite any post-sale restriction) exhausts a patentee's rights, leaving only contract remedies.
A unanimous Supreme Court held that method patents are subject to exhaustion and that an authorized sale of components substantially embodying a patent ends the patentee's rights.
The Supreme Court held that a good-faith belief in a patent's invalidity does not negate the intent required for induced infringement, while reaffirming that inducement demands knowledge of infringement.
The Supreme Court abolished licensee estoppel, holding that federal patent policy lets a licensee stop paying royalties and challenge the validity of the licensed patent.
The Supreme Court held that inducing patent infringement under § 271(b) requires knowledge of infringement, but that willful blindness to a patent can supply that knowledge.
A unanimous Supreme Court held that inducing infringement under Section 271(b) requires a predicate act of direct infringement, refusing to extend liability across divided method steps.
The Supreme Court rejected an absolute bar on equivalents after a narrowing amendment, replacing it with a rebuttable presumption that still governs the doctrine of equivalents today.
