Skip to content
PatentBrief

Technology Patents

Computer-Implemented Invention Patents

USPTO Alice two-step framework, EPO technical character test, and global CII claim drafting strategy for software and algorithm patents.

FAQ

What is the USPTO framework for patenting computer-implemented inventions?

The USPTO applies the Alice/Mayo two-step framework (codified in the January 2019 Revised Guidance) to evaluate computer-implemented inventions under 35 U.S.C. § 101: THE ALICE FRAMEWORK: STEP 1 (MAYO STEP ONE): Is the claim directed to a process, machine, manufacture, or composition of matter? If yes, proceed to Step 2A; if no, § 101 satisfied. STEP 2A PRONG 1: Does the claim recite an abstract idea? Abstract idea categories: (1) mathematical concepts (mathematical relationships; mathematical formulas; mathematical calculations; mathematical algorithms); (2) certain methods of organizing human activity (fundamental economic principles or practices; commercial or legal interactions; managing personal behavior or relationships); (3) mental processes (concepts performed in the human mind including observation; evaluation; judgment; opinion). STEP 2A PRONG 2 — PRACTICAL APPLICATION: even if the claim recites an abstract idea, does it integrate the abstract idea into a practical application? Integration indicators: (a) improvement in the functioning of a computer or another technology — this is the Enfish principle; (b) application to a particular machine that is integral to the claim; (c) effecting a transformation of a particular article; (d) applying an abstract idea in some other meaningful way beyond generally linking it to a technology environment. STEP 2B — INVENTIVE CONCEPT: if no practical application under Prong 2, do the additional claim elements add significantly more? Individual elements: routine/conventional computer operations like storing, retrieving, outputting information typically do NOT add significantly more; unconventional combinations: even routine elements in an unconventional arrangement CAN add significantly more (Berkheimer v. HP, Fed. Cir. 2018: whether elements are well-understood is a question of FACT, not law). KEY CASES: ELIGIBLE: Enfish (self-referential table that improved computer memory efficiency); McRO (specific rules-based process for animated lip sync that a human could not practically perform mentally); Core Wireless (specific user interface improvement improving mobile device display); INELIGIBLE: Alice Corp v. CLS Bank (intermediated settlement on a computer); Electric Power Group (collecting and analyzing data from power grid); BSG Tech (applying relational model database to prior art method).

How does the EPO evaluate computer-implemented inventions?

The European Patent Office has a fundamentally different but practically effective approach to computer-implemented inventions that avoids the US abstract idea problem in most cases: THE EPO TECHNICAL CHARACTER APPROACH: ARTICLE 52 EPC: programs for computers are excluded from patentability as such; but 'as such' is the operative phrase — the EPO has substantially narrowed this exclusion; COMVIK DECISION (T 641/00): established the framework for mixed technical/non-technical inventions; a claim is not excluded if it has technical character; even if non-technical features are present, they can contribute to the claim's inventive step IF they serve a technical purpose; AT&T KNOWLEDGE VENTURES (T 1784/06): computer programs are patentable if they cause a 'further technical effect' beyond the normal physical interactions between a program and the computer on which it runs; IBM COMPUTER PROGRAMS DECISIONS: T 1173/97 and T 935/97: established that a computer program product is not excluded from patentability if it produces a technical effect when run on a computer; WHAT COUNTS AS TECHNICAL: (1) improving computer security (preventing unauthorized access; detecting intrusions; encrypting data); (2) improving computer performance (faster processing; reduced memory use; optimized network communication); (3) improving physical devices controlled by software (medical devices; industrial automation; vehicle control systems); (4) data compression algorithms (reduce actual storage space or transmission bandwidth); (5) signal processing algorithms (process physical signals); WHAT IS NOT TECHNICAL: (1) purely economic/financial schemes; (2) presentations of information solely for their cognitive content; (3) mathematical algorithms with no technical application; CLAIM FORMAT FOR EPO: system + method + computer program product triple claims are standard; the computer program product claim ('A computer program comprising instructions which, when executed by a processor, cause the processor to perform...') is fully patentable at EPO if the method it implements has technical character; PROSECUTION TIPS: never characterize the problem to be solved in purely business/economic terms; always frame the problem in technical terms (improve processing efficiency; reduce network latency; prevent security attacks); the technical character requirement gives more latitude than Alice, but claims must consistently use technical framing.

How should CII claims be drafted for global protection?

A well-drafted computer-implemented invention claim can often be drafted in a single format that satisfies both the USPTO and EPO requirements simultaneously, enabling cost-efficient global protection: THE UNIVERSAL CLAIM FORMAT: SYSTEM CLAIM: 'A computing system comprising: at least one processor; a memory storing instructions that, when executed by the at least one processor, cause the computing system to: [specific technical steps with technical terminology]'; METHOD CLAIM: 'A computer-implemented method comprising: [specific technical steps, each described with technical precision]'; CRM CLAIM: 'A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the processors to: [same technical steps]'; WHAT MAKES A CLAIM BOTH EPO AND USPTO FRIENDLY: (1) USE TECHNICAL LANGUAGE FOR THE PROBLEM: 'to reduce computational complexity from O(n²) to O(n log n)' rather than 'to enable faster analysis'; 'to reduce network round-trips by implementing server-side state caching' rather than 'to improve user experience'; (2) SPECIFIC ARCHITECTURE: reference specific technical components (hash tables; decision trees; neural network layers; message queues; cache eviction policies); (3) FUNCTIONAL LANGUAGE TIED TO TECHNICAL RESULTS: 'wherein the compression ratio exceeds that achievable by [prior art method] by at least [X]% for [specific data type]'; (4) MEASUREMENTS: if the invention provides quantifiable technical improvements, include them in the dependent claims and emphasize them in prosecution; CLAIM HIERARCHY STRATEGY: BROAD INDEPENDENT CLAIM: functional description of the technical improvement; capture the broadest non-abstract formulation; INTERMEDIATE DEPENDENT CLAIMS: specific algorithms; specific data structures; specific protocol interactions; NARROW DEPENDENT CLAIMS: specific implementation choices; specific parameter ranges; SPECIFICATION REQUIREMENTS: pseudocode or flowcharts showing the algorithm in detail; data structure diagrams; test data or benchmarks; comparison to prior art performance; description of specific machine interactions; CLAIM DRAFTING MISTAKES TO AVOID: claiming the RESULT without the TECHNICAL MEANS ('a method for improving customer satisfaction' vs. 'a method comprising specific technical steps'); using only business/economic language even in technical claims ('optimize revenue' rather than 'reduce processing latency'); claims that read as 'apply [abstract idea] on a computer' without specifying HOW the computer is used differently.

How do other major jurisdictions — China, Japan, and Korea — handle CII patents?

Each major Asian patent jurisdiction has developed its own approach to computer-implemented inventions that requires specific claim drafting adaptations: CHINA — CNIPA APPROACH: GENERAL RULE: software 'per se' is excluded from patent protection under Article 25 of the Chinese Patent Law; BUT: CIIs that constitute 'technical solutions' are patentable; KEY TEST: does the claim relate to a method or apparatus that solves a technical problem using technical means to achieve a technical effect?: technical means = specific algorithm or data processing steps; technical problem = a problem in the technical field; technical effect = a tangible, measurable improvement in the technical field; DRAFTING FOR CHINA: claims must explicitly name the processor, memory, or specific hardware; software steps must be tied to hardware; data fields in database claims should reference what type of data and why it solves a technical problem; EXAMPLES: ELIGIBLE (CHINA): a method for reducing database query time using a specific B-tree index variant that reduces query complexity from O(n) to O(log n); INELIGIBLE (CHINA): a method for managing customer relationship data; JAPAN — JPO APPROACH: Article 2(3)(i) of the Japan Patent Act: 'creation of technical ideas utilizing natural laws'; COMPUTER PROGRAMS: computer programs constitute 'things' that can be patented in Japan; programs = creation of technical ideas using natural laws as long as information processing by the software is concretely realized using hardware; SOFTWARE PATENT ELIGIBILITY: software must process concrete information using hardware; the software must exploit the hardware in a technical way; DRAFTING FOR JAPAN: claims should describe the software in terms of what hardware functions it controls; 'a computer program that causes a processor to execute the steps of' is acceptable; KOREA — KIPO APPROACH: similar to Japan in requiring that information processing be performed using hardware; Korean IP Office examination guidelines closely follow the EPO approach; claims must have technical character; 'computer programs recorded on computer-readable media' are patentable; PATENT COOPERATION TREATY (PCT) STRATEGY: use EPO/USPTO-compatible claim format as the base; add Japan/China/Korea specific claim sets in national phase entry if needed; the EPO technical character framing is generally the most universally acceptable baseline.

Related Guides

Alice § 101 FrameworkSoftware Patent StrategyAI/ML PatentsBlockchain Patents