The numbered sentences at the end of a patent that legally define the scope of the protected invention. Governed by 35 U.S.C. § 112(b), each claim must particularly point out and distinctly claim the invention, setting the precise boundary of what infringes the patent.
Last reviewed: 2026/06/17
The process by which a court determines the meaning and scope of the terms in a patent claim. In US litigation, claim construction is a question of law decided by the judge — typically at a 'Markman hearing' — and it governs how the patent reads on accused products and prior art.
Legal PracticeThe process of drafting, filing, and negotiating a patent application before a patent office such as the USPTO — from initial filing through examination, office actions, and responses to grant or abandonment. Distinct from patent litigation, which enforces issued patents in court.
Legal PracticeWhether an invention meets the statutory requirements for a U.S. patent: patent-eligible subject matter and utility (35 U.S.C. §101), novelty (§102), non-obviousness (§103), and the §112 disclosure requirements of written description, enablement, and claim definiteness.
Legal PracticePrior art is the body of publicly available knowledge — patents, published applications, printed publications, public use, on-sale activity, and anything otherwise available to the public before an invention's effective filing date — used to assess whether a claimed invention is new and non-obvious.
Move from this definition to role-based legal AI shortlists and the selection criteria that matter for each type of legal team.
Specialist firm workflows: deep practice area expertise, premium client service, selective tool adoption.
Last reviewed: 2026/06/17. Definitions are written by the LawyerAI Editorial team. Commercial relationships are disclosed and do not determine editorial scores or conclusions. See our Sponsorship & Affiliate Disclosure.
A patent claim is one of the numbered sentences at the end of a patent that legally defines the scope of the protected invention. Governed in the United States by 35 U.S.C. § 112(b), each claim must "particularly point out and distinctly claim" the subject matter the inventor regards as the invention. The claims — not the drawings, the abstract, or the descriptive specification — determine exactly what the patent owner can stop others from making, using, selling, or importing.
The claims are the operative legal text of a patent. Everything else in the document — the background, the detailed description, the figures — exists to support and explain the claims, but it is the claims that define the metes and bounds of the property right. For a litigator, infringement is decided claim by claim: an accused product infringes only if it contains every limitation of at least one asserted claim. For a prosecutor, the central task is to obtain claims that are broad enough to be commercially valuable yet narrow enough to be allowed over the prior art and to survive a validity challenge.
This dual pressure makes claim drafting one of the highest-leverage activities in legal practice. A single misplaced word can render a claim either too narrow to deter competitors or so broad that it reads on the prior art and is invalid. Claims that fail the definiteness requirement of § 112(b) — those that do not inform a person skilled in the art, with reasonable certainty, of the scope of the invention — can be held invalid even years after issuance.
Claims also drive valuation and transactions. When a portfolio is licensed, acquired, or asserted, sophisticated parties read the independent claims first to understand what is actually protected. Freedom-to-operate analysis, design-around strategy, and infringement opinions all begin with a careful reading of the claims, which is why claim literacy is foundational for litigators, prosecutors, and transactional counsel alike.
Anatomy of a claim. A patent claim is, grammatically, a single sentence with three parts:
Elements and limitations. Each distinct feature recited in the body is a claim element or limitation. The "all-elements rule" governs infringement: an accused product or process must contain each and every limitation, either literally or under the doctrine of equivalents. Adding a limitation narrows a claim; removing one broadens it.
Independent versus dependent claims. An independent claim stands on its own and recites a complete set of limitations. A dependent claim refers back to an earlier claim ("The method of claim 1, further comprising...") and incorporates all of that claim's limitations, then adds at least one more. A dependent claim is therefore always narrower than the claim it depends from. Drafters commonly pair a few broad independent claims with a series of dependent claims that supply progressively narrower fallback positions in case the broadest claims are found invalid.
Means-plus-function limitations. Under 35 U.S.C. § 112(f), an applicant may express a claim element "as a means or step for performing a specified function without the recital of structure." When a limitation is interpreted under § 112(f), its scope is limited to the corresponding structure, material, or acts disclosed in the specification and their equivalents — not to any structure that performs the function. The Federal Circuit in Williamson v. Citrix Online (2015) eliminated the strong presumption that the word "means" was required to invoke § 112(f), so functional language using generic placeholders ("module for," "mechanism for") can trigger means-plus-function treatment. Drafters must therefore choose functional language deliberately, either to invoke or to avoid § 112(f).
Claim differentiation. Because each claim is presumed to have a different scope, the presence of a particular limitation in a dependent claim implies that the broader independent claim does not contain that limitation. This canon of claim differentiation is frequently used during claim construction to argue for a broader reading of an independent claim.
Claim scope is a strategic decision, not a clerical one. The breadth of the independent claims sets the commercial value of the patent. Firms should align claim strategy with the client's business objectives — what competitors are likely to build, where design-arounds are feasible, and which features are commercially essential.
Definiteness must be tested against § 112(b). Every claim term should be checked for whether it conveys scope with reasonable certainty to a person skilled in the art. Relative terms ("substantially," "about," "near") are permissible only when the specification or the art supplies an objective benchmark. Antecedent-basis errors — using "the widget" before "a widget" has been introduced — are a common and avoidable source of indefiniteness rejections.
Functional language requires deliberate handling. Because § 112(f) can be triggered without the word "means," firms should adopt drafting standards that make the intended interpretation explicit and ensure the specification discloses adequate corresponding structure, particularly for computer-implemented inventions where an algorithm may be required.
Support and antecedent basis must trace to the specification. Claim amendments made during patent prosecution must find written-description support in the application as filed. Firms should confirm that every claimed limitation is described, not merely mentioned, before filing.
Prosecution history is permanent. Arguments and amendments made to overcome rejections can give rise to prosecution-history estoppel, narrowing the scope available under the doctrine of equivalents. Statements made about claim meaning during prosecution will be read back against the patent owner in litigation.
Broad claims invite invalidity. A claim broad enough to cover many competitor products is also more likely to read on the prior art and be invalidated for anticipation or obviousness, or to fail § 112 for lack of enablement or written description. The tension between breadth and validity is inherent and cannot be drafted away entirely.
Indefiniteness can void a claim entirely. A claim that fails to inform skilled artisans of its scope with reasonable certainty is invalid under § 112(b). Indefiniteness is a question that can be raised at the USPTO during examination, at the Patent Trial and Appeal Board, and in district court litigation.
Claim construction is unpredictable. The meaning of a disputed term is decided by the court (in the United States, in a Markman hearing), and reasonable judges can reach different constructions of the same language. A construction that the patent owner did not anticipate can convert a strong infringement position into a non-infringement or invalidity finding.
Jurisdictional variation. Claim format and interpretation differ across patent offices. European Patent Office practice favors the two-part "characterized in that" form; other jurisdictions limit claim counts or apply distinct doctrines of equivalents. A claim set drafted for the USPTO often requires adaptation for foreign filings, and counsel should not assume US claim interpretation principles transfer directly abroad.
AI assistance does not transfer responsibility. Tools that generate or analyze claim language accelerate the work but do not assume legal responsibility for scope, definiteness, or validity. A registered patent attorney or agent must review all machine-generated claim language, and unauthorized practice rules apply to claim drafting regardless of the tools used.