DETAILED ACTION
Response received on May 20, 2026 has been acknowledged. Claims 1, 12-14, and 19-20 have been amended. Therefore, Claims 1-20 are pending.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of Claims
This Non-Final Office action is in response to the application filed on 05/21/2024 and in response to Applicant’s Arguments/Remarks filed on 05/20/2026. Claims 1-20 are pending.
Priority
Application 18/670,155 was filed on 05/21/2024.
Applicant’s Reply
Applicant's response of May 20, 2026 has been entered. The examiner will address applicant’s remarks at the end of this office action. The examiner acknowledges the amendments made to Claim 1, 12-14, and 19-20.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 5/20/2026 has been entered.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Claims 1‐13 are directed to a system (machine), Claims 14-18 are directed to a method (process), and Claims 19-20 are directed to a non-transitory computer readable storage medium storing (machine/apparatus). Thus, these claims fall within one of the four statutory categories of invention. (Step 1: YES).
For step 2A, the Examiner has identified independent method Claim 1 as the claim that represents the claimed invention for analysis and is similar to independent claim 14 and 19. Claim 1, as exemplary is recited below, isolating the abstract idea from the additional elements, wherein the abstract idea is set in bold:
An electronic online system for compliance analysis of an organization, the online system comprising: a processor subsystem; and memory including instructions, which when executed by the processor subsystem, cause the processor subsystem to: receive, from a user of the electronic online system, an indication of a law for analysis; parse the law to produce law chunks, wherein the parsing produces law chunks sized based on a token limit of a context window for a large language model (LLM); provide the law chunks to the large language model (LLM) with a prompt to determine whether each law chunk is a business-relevant law chunk that is relevant to the organization, a business line of the organization, a product line of the organization, a business area of the organization, or a market area, and to discard law chunks determined to be not relevant, wherein the large language model analyzes the law chunks in the context window and discards irrelevant law chunks, wherein discarding chunks determined to be not relevant reduces a token count of inputs; reparse the law to produce more granular business-relevant law chunks of the business-relevant law chunks, a chunk size of the reparsed business-relevant law chunks determined by an amount of available token space in the context window; receive, from the user, an indication of a business policy for analysis; parse the business policy to produce policy chunks at a granularity similar to the reparsed business-relevant law chunks; compare the reparsed business-relevant law chunks with the policy chunks to determine similarity scores for respective pairs of law chunks and policy chunks; and present law chunks that have similarity scores less than a threshold similarity score to the user.
The above bolded limitations recite the abstract idea of providing corporate compliance analysis configured to find the gaps in a company’s coverage of laws and regulations that are applicable to it. These limitations under its broadest reasonable interpretation, covers certain methods of organizing human activity (i.e., commercial or legal interactions including agreements in the form of contracts, legal obligations, advertising, marketing or sales activities or behaviors, and business relations as well as fundamental economic principles) but for the recitation of generic computer components. That is, other than reciting a system implemented by a data processor (computer) the claimed invention amounts to the abstract idea stated above. For example, for the related computer components, this claim encompasses legal actions that could conventionally be performed by compliance officers, lawyers, or business auditors manually as part of company compliance or auditing process. This progress can be done manually through paper and pencil. Additionally, providing corporate compliance analysis is considered a commercial or legal interactions because it involves compliance review between laws and business policies, which are basic methods of organizing human activity related to commercial and regulatory decision-making. If a claim limitation, under its broadest reasonable interpretation, covers commercial or legal interactions between parties, but for the recitation of generic computer components, then it falls within the “certain methods of organizing human activity” grouping of abstract ideas. The mere nominal recitation of a “an electronic online system”, “a processor subsystem”, “a memory”, and “a token limit of a context window for a large language model (LLM)”, “a large language model LLM”, and “available token space in the context window” do not take the claim out of the methods of organizing human interactions grouping. Thus, claims 1, 14, and 19 recites an abstract idea. (Step 2A- Prong 1: YES. The claims recite an abstract idea).
This judicial exception is not integrated into a practical application (2nd prong of eligibility test for step 2A). In particular, Claim 1 recites additional elements of “an electronic online system”, “a processor subsystem”, “a memory”, and “a token limit of a context window for a large language model (LLM)”, “a large language model LLM”, and “available token space in the context window”. Claim 14 recites the additional elements of “an electronic online system”, a token limit of a context window for a large language model (LLM)”, “a large language model LLM”, and “available token space in the context window” Claim 19 recites additional elements of “non-transitory machine-readable medium”, “an electronic online system”, a token limit of a context window for a large language model (LLM)”, “a large language model LLM”, and “available token space in the context window”. These additional elements are all considered nothing more than generic computing devices to perform generic communicating functions such as storing data and instructions, transmitting and receiving data between computers. Further, the recited LLM is merely generic implementation of machine learning technology used as a tool to perform the claimed information analysis, and the claims do not recite any technological improvement to the underlying machine learning model, its architecture, training, or operation. Accordingly, these additional elements (combination of computer and the use ledgers) do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea and are recited at a high level of generality when considered both individually and as a whole. Thus, Claims 1, 14, and 19 are directed to an abstract idea without an integration into a practical application. (Step 2A-Prong 2: NO: the additional claimed elements are not integrated into a practical application).
For step 2B, the claim(s) do not include additional elements that are sufficient to amount to significantly more than the judicial exception because they do not amount to more than simply instructing one to practice the abstract idea by using generic computer components to carry out the steps that define the abstract idea, as discussed above. This does not render the claims as being eligible. See MPEP 2106.05(f). The additional elements of using an “an electronic online system”, “a processor subsystem”, “a memory”, and “a token limit of a context window for a large language model (LLM)”, “a large language model LLM”, and “available token space in the context window” when considered both individually and as an ordered combination did not add significantly more to the abstract idea because they were simply applying the abstract idea using generic computer components. The recited token limit and available token space merely represent generic operating parameters of an LLM and do not improve the functionality of the underlying computer or machine learning technology. In addition, the claims recite the additional elements which are considered nothing more than a general link to technology because there is no recitation of specifics of how this additional element is being used. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept (See MPEP 2106.05(f)). Accordingly, these additional elements, do not change the outcome of the analysis, and claims 1, 14, and 19 are not patent eligible. (Step 2B: NO. The claims do not provide significantly more).
Claims 2-4, and 15 recite limitations that further define the same abstract idea of independent claims to include wherein the indication of the law includes a filename of a document that includes the law, wherein the indication of the law includes a universal resource locator (URL) of a document that includes the law, and wherein the indication of the law includes a database identifier of a record that includes the law. The claimed elements are considered part of the abstract idea because they merely define how to provide corporate compliance analysis configured to find the gaps in a company’s coverage of laws and regulations, without adding any concrete technological implementation or improvement. Additionally, the dependent claim does not include any new additional elements and therefore are considered patent ineligible for the reasons given above.
Claims 7-9 and 17 recite limitations that further define the same abstract idea of independent claims to include wherein the indication of the business policy includes a filename of a document that includes the business policy, wherein the indication of the business policy a universal resource locator (URL) of a document that includes the business policy, and wherein the indication of the business policy includes a database identifier of a record that includes the business policy. The claimed elements are considered part of the abstract idea because they merely define how to provide corporate compliance analysis configured to find the gaps in a company’s coverage of laws and regulations, without adding any concrete technological implementation or improvement. Additionally, the dependent claim does not include any new additional elements and therefore are considered patent ineligible for the reasons given above.
Claim 5 and 10 recite limitations that further define the same abstract idea of independent claims to include wherein parse the law to produce law chunks, apply a chunking algorithm to the law and wherein to parse the business policy to produce policy chunks, apply a chunking algorithm to the business policy. In addition, claim 5 and 10 recite the additional element of “the processor subsystem” which is considered nothing more than a general link to a technological environment because there is no recitation of specifics of how this additional element is being used. See MPEP 2106.05(f) and (h) indicate that merely “generally linking” the abstract idea to a particular technological environment or field of use cannot provide a practical application or significantly more. Therefore claim 5 and 10 are patent ineligible.
Claims 6, 11, 16, and 18, recite limitations that further define the same abstract idea of independent claims to include wherein the chunking algorithm includes at least one of: naïve splitting, recursive chunking, or semantic chunking. In addition, claim 6,11,16, and 18 recite the additional element “a sentence-level tokenizer process” and “a sentence-level tokenizer process with context preservation” which is considered nothing more than a general link to a technological environment because there is no recitation of specifics of how this additional element is being used. See MPEP 2106.05(f) and (h) indicate that merely “generally linking” the abstract idea to a particular technological environment or field of use cannot provide a practical application or significantly more. Therefore claim 6, 11, 16, and 18 are patent ineligible.
Claims 12-13, and 20 recite limitations that further define the same abstract idea of independent claims to include wherein calculate a vector representation of a law chunk; calculate a vector representation of a policy chunk; use a vector comparison operation to compare the vector representation of the law chunk to the vector representation of the policy chunk, the vector comparison producing a similarity score, and wherein the vector comparison is one of: a dot product operation, a cosine similarity operation, or a soft cosine similarity operation. The claimed elements are considered part of the abstract idea because they merely define how to provide corporate compliance analysis configured to find the gaps in a company’s coverage of laws and regulations, without adding any concrete technological implementation or improvement. Additionally, the dependent claim does not include any new additional elements and therefore are considered patent ineligible for the reasons given above.
Subject Matter Free of Prior Art
The prior art of record neither anticipates nor supports a conclusion of obviousness without the use of impermissible hindsight with respect to the subject matter which is present in independent claims 1, 14, and 19.
In regards to Claims 1,14, and 19 with respect to the prior art, the closest reference appears to be Hoang et al. (US20230237399) in view of Chen et al. (US 20250182138), in view of Massie et al. (US 20250272581), further in view of PONDICHERRY MURUGAPPAN et al. (US20200111023):
Hoang et al. teaches a processor subsystem; (See [0003]) a memory including instructions, which when executed by the processor subsystem, cause the processor subsystem to (See [0053]): parse the law to produce law chunks; (See [0021]) parse the business policy to produce policy chunks; (See [0019], [0021]- [0022]); compare the business-relevant law chunks with the policy chunks to determine similarity scores for respective pairs of law chunks and policy chunks; (See [0093], [0099]-[0100]) provide the law chunks to a large language model LLM with a prompt to determine whether each law chunk is a business-relevant law chunk that is relevant to the organization, a business line of the organization, a product line of the organization, a business area of the organization, or a market area, (See [0016], [0125], [0129]). Chen et al. teaches receive, from a user of the electronic online system, an indication of a law for analysis; (See [0011], [0030], [0039],) receive, from the user, an indication of a business policy for analysis; (See [0011], [0024]). Massie et al. teaches discard [law] chunks determined to be not relevant; (See [0094].). PONDICHERRY MURUGAPPAN et al. teaches present [law chunks] that have similarity scores less than a threshold similarity score to the user (See [0051]).
However, Hoang et al. in view of Chen et al., in view of Massie et al., further in view of PONDICHERRY MURUGAPPAN et al. alone or in combination fail to disclose or render obvious parse the law to produce law chunks, wherein the parsing produces law chunks sized based on a token limit of a context window for a large language model (LLM); to discard law chunks determined to be not relevant, wherein the large language model analyzes the law chunks in the context window and discards irrelevant law chunks, wherein discarding chunks determined to be not relevant reduces a token count of inputs; reparse the law to produce more granular business-relevant law chunks of the business-relevant law chunks, a chunk size of the reparsed business-relevant law chunks determined by an amount of available token space in the context window; and parse the business policy to produce policy chunks at a granularity similar to the reparsed business-relevant law chunks, when the claim is considered as a whole. Absent a suggestion or teaching in the prior art, the examiner will not engage in impermissible hindsight to supply the missing limitation(s). Therefore, independent claim 1, 14, and 19 and their dependent claims, are not rejected under prior art.
Response to arguments
Applicant's arguments filed 5/20/2026 have been fully considered but they are not persuasive.
The comments regarding the 35 USC 101 rejection are noted. On page 8 of Applicant’s response, applicant asserts that the claims are not directed to an abstract idea (Step 2A, Prong One) and that the Examiner characterizes the claims as directed to "providing corporate compliance analysis configured to find the gaps in a company's coverage of laws and regulations.” and submits that the amended claims recite specific technical operations that are not performable in the human mind or through pen-and-paper methods. The Examiner respectfully disagrees because when considered as a whole, the claims are directed to evaluating legal requirements against an organization’s business policies to identify compliance gaps, which constitutes a commercial legal interaction involving the management of legal obligations between an organization and governing regulatory frameworks, and the recited LLM based processing merely services as a toll for performing that abstract analysis. Applicant further argues that as amended, claim 1 recites operations are rooted in the computational constraints of LLM architectures specifically, context window token limits and involve iterative parsing calibrated to available token space and that no human actor performs "reparsing" based on "available token space in the context window" of a language model. These are machine-specific operations without a mental-process analog. The Examiner respectfully disagrees because the recited context-window token limits, available token space, and other additional elements merely define the environment in which the claimed compliance analysis is performed and do not alter the nature of the claims, which remain directed to evaluating laws and policies to determine regulatory compliance, a commercial legal interaction.
Applicant further argues that even assuming arguendo that some aspect of the claims touches upon compliance analysis, the amended claims integrate that concept into a practical application by reciting a specific technical solution to a problem arising in the LLM computing environment. The Examiner respectfully disagrees because the recited LLM-based chunking, token-limit management, and reparsing operations merely facilitate the performance of the claimed compliance analysis using generic machine learning technology and do not improve the operation of the LLM, the context-window architecture, or any other computer technology, such that the claims fail to integrate the judicial exception into a practical application. Applicant further argues that the problem: LLMs have finite context windows measured in tokens. Processing an entire body of law in a single pass exceeds these limits. The solution recited in the claims: parse the law into chunks sized to the context window; filter irrelevant chunks via the LLM to reduce token count; then reparse the remaining relevant chunks at a finer granularity determined by the freed token space. This iterative, token-aware chunking-and-filtering pipeline is a technical improvement to how documents are processed in LLM-based systems. The Examiner respectfully disagrees because the recited token-aware chunking, filtering, and reparsing operations merely organize and prepare information for analysis using generic operating parameters of an LLM and do not improve the functionality, architecture, or underlying operation of the LLM or the other computer technology, but instead use the LLM as a tool to perform the claimed commercial legal compliance analysis.
Applicant further argues that here, the claims improve LLM-based document processing by dynamically managing token allocation across parsing stages. The Examiner respectfully disagrees because the alleged improvement is directed to the manner in which information is organized, filtered, and prepared for analysis rather than to an improvement in the operation of the LMM itself or other computer technology. While the claims recite dynamically managing token allocation across parsing stages, such operations merely utilize generic LLM constraints and parameters to facilitate the underlying abstract idea and do not improve the architecture, training, functionality, or performance of the LLM. Accordingly, the claims use machine learning technology as a tool to perform the recited commercial legal interaction rather than integrating the judicial exception into a practical application. Applicant further argues that The Examiner's characterization of the LLM as a "general link to technology" is inapplicable to the amended claims. The claims do not merely invoke an LLM generically; they recite a multi- stage pipeline in which chunk sizing, relevance filtering, token-count reduction, and reparsing are each constrained by the LLM's context window capacity. This is a specific implementation, not a generic invocation. The Examiner respectfully disagrees because although the claims recite multi-step process constrained by a LLM’s context window capacity, the recited chunk sizing, relevance filtering, token-count reduction, and reparsing operations merely organize and manage information using generic operating parameters of an LLM and do not improve the underlying architecture or functionality of the LMM or other computer technology, Rather, the LLM is used as a tool to facilitate the claimed commercial legal compliance analysis. Applicant further argues that no evidence has been presented, nor does the record support, that this particular ordered combination was well- understood, routine, or conventional. The Examiner notes that the Applicant’s argument is not persuasive because the Office Action did not reject the claims on the basis that the ordered combination of elements was well- understood, routine, or conventional. Thus, the applicants argument does not address the basis of the rejection. Thus, the claims do not amount to a technological improvement and fail to provide an inventive concept. The argument to the contrary is not persuasive. Thus, the rejections of Claims 1-20 under 35 USC 101 are maintained.
The comments regarding the 35 USC 103 rejection are noted. On page 10 of Applicant’s response, applicant asserts the cited art, individually or in combination, fails to disclose or suggest the combination of features now recited in the dependent claims. Examiner finds the arguments persuasive. The prior art of record alone or in combination fail to disclose or render obvious parse the law to produce law chunks, wherein the parsing produces law chunks sized based on a token limit of a context window for a large language model (LLM); to discard law chunks determined to be not relevant, wherein the large language model analyzes the law chunks in the context window and discards irrelevant law chunks, wherein discarding chunks determined to be not relevant reduces a token count of inputs; reparse the law to produce more granular business-relevant law chunks of the business-relevant law chunks, a chunk size of the reparsed business-relevant law chunks determined by an amount of available token space in the context window; and parse the business policy to produce policy chunks at a granularity similar to the reparsed business-relevant law chunks, when the claim is considered as a whole. Absent a suggestion or teaching in the prior art, the examiner will not engage in impermissible hindsight to supply the missing limitation(s). Thus, the rejection of claims 1-20 under 35 U.S.C. §103 has been withdrawn.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALAA WADIE HUSSEIN whose telephone number is 571-270-1748. The examiner can normally be reached M-F: 8:00-5:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jessica Lemieux can be reached on 571-270-3445. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/A.W.H./
Examiner, Art Unit 3626
/JESSICA LEMIEUX/Supervisory Patent Examiner, Art Unit 3626
Prosecution Insights