DETAILED ACTION
This communication is in response to the Amendments and Arguments filed on 02/06/2026.
Claims 1, 3-9, 11-17, 19, and 20 are pending and have been examined.
All previous objections/rejections not mentioned in this Office Action have been withdrawn by the examiner.
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 .
Response to Arguments
Applicant's arguments filed 02/06/2026 have been fully considered but they are not persuasive.
Regarding the 101 rejections, Applicant asserts on pgs 6-9 that the claim limitations are inherently tied to computing systems and cannot practically be performed in the human mind, and further describe a technical solution to a technical problem beyond merely reciting using AI or generating code. The Examiner respectfully disagrees with these assertions. A human programmer with a learned understanding of natural language as well as computer language is capable of hearing a natural language request for a computer code that can perform a specific function when executed, determining additional requirements or boundary conditions the completed code must satisfy, and using their learned understanding to write out a code in specific formats and sections, where the code is capable of being compiled and executed if it were to be properly loaded into a computer system. Validating the file reads to the human programmer proofreading their code to identify typographical errors, incorrectly used formulas, calling the incorrect number of arguments, or the like, to ensure that the code would function as intended. The Examiner notes that the independent claims recite “the one or more server files being compilable and executable in the framework server”, which is not an active recitation of compiling and executing the files. Even dependent claims 7 and 15, which recite “attempting to compile…” can be interpreted as a programmer ensuring the code is in a machine readable format and checking to make sure it can be compiled. Under BRI, the attempt includes validating the code prior to compiling, as there is no recitation of a successful/completed compiling of the files. The technological improvements asserted by the Applicant on pg 8 related to improving reliability and correctness of generated server implementations are not clearly present in the claim language, as there is nothing to indicate the files are improved after failing validation – only that they are re-generated (see claim 6, which ends with retrying the request without an indication that the output of retrying is improved). While the specification may detail an improvement in a technology or technical field, the claims recite generalized computer components that are performing the processes recited in the claims. As per MPEP 2106.05(a)(II) and 2106.05(f), using generalized computer components to perform the recited process amounts to mere instructions to apply an exception using a generic computer component, and the claim must include more than mere instructions to perform the method on a generic component or machinery to qualify as an improvement to an existing technology. Therefore, the claims are still directed to an abstract idea without additional elements that are sufficient to amount to significantly more than the judicial exception. Regarding the arguments related to the Desjardins decision, the decision itself is specifically addressing a claim set that recites an improvement to the way a machine learning model is trained. In the instant application, the claims do not recite any training of the GAI model, nor do they indicate any improvement to the GAI model itself. Therefore, the Desjardins decision does not have any direct correlation to the 101 interpretation of the instant claims. The claims remain patent ineligible.
Regarding the 103 rejections, Applicant asserts on pg 11 that Jiang does not teach server files in a declarative format, and further does not teach the amended claim language. The Examiner respectfully disagrees with these assertions. There is no recited detail for what a “declarative-based data model definition” actually comprises, nor what a “server file” or “software-based server service” are. Using the BRI for these terms Jiang in view of Wang teaches the amended claims. Jiang teaches providing a response to a prompt that can include natural language output, such as using a fallback prompt resulting in model-generated suggestions for 3 components to an overall code (see Appendix B.1), or can produce HTML or JavaScript (see Intro), which reads to a declarative-based data model definition. Jiang further teaches the resulting code is used for apps or buttons as part of human computer interactions and computing, which read to a server file and software-based server service (see Fig. 1, CCS Concepts, Abstract, Intro), and Wang further teaches a server system (see [0030]). Wang further teaches that code projects are compiled and packaged (see [0057]). Therefore, the combination of cited art still teaches the amended claims. Please see the updated mappings below for further detail.
Hence, Applicant’s arguments are not persuasive.
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, 3-9, 11-17, 19, and 20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding claim(s) 1, 9, and 17, the limitation(s) of receiving a text-based prompt, creating a request, feeding the request, and storing, as drafted, are processes that, under broadest reasonable interpretation, covers performance of the limitation in the mind and/or with pen and paper but for the recitation of generic computer components. More specifically, the mental process of a human programmer reading a description of a process, adding other related information to the description, using the combined information and learned rules for how to develop process steps to write out a set of steps in the form of a computer code, and setting it aside for future use. The generative artificial intelligence model reads to a set of learned rules for how to analyze, process, and produce a result using human language. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind and/or with pen and paper but for the recitation of generic computer components, then it falls within the --Mental Processes-- grouping of abstract ideas. Accordingly, the claim(s) recite(s) an abstract idea.
This judicial exception is not integrated into a practical application because the recitation of a system, repository, server, and framework client in claim 1, framework client, framework server, and server repository in claim 9, and machine-readable medium, processors, framework client, framework server, and server repository in claim 17, reads to generalized computer components, based upon the claim interpretation wherein the structure is interpreted using [0082-91] in the specification. Accordingly, these additional elements do not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claim(s) is/are directed to an abstract idea.
The claim(s) do(es) not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element of using generalized computer components to receive, create, feed, and store, amounts to no more than mere instructions to apply the exception using a generic computer component. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept. The claim(s) is/are not patent eligible.
With respect to claim(s) 3, 4, 11, 12, 19, and 20 the claim(s) recite(s) features of the files, which reads on a human writing out the steps in a specific format and/or code language and with specific additional information and/or features. No additional limitations are present.
With respect to claim(s) 5 and 13, the claim(s) recite(s) creating, feeding, and storing, which reads on a human using information from the previously determined code information as well as a second set of related information to determine and write out additional information using learned rules, and filing all the papers with pertinent information related to the code together. No additional limitations are present.
With respect to claim(s) 6 and 14, the claim(s) recite(s) validating and retrying, which reads on a human reviewing/proofreading the code to determine if it will work correctly, and rewriting the code using the description, related information, and learned rules if it would not work as intended. No additional limitations are present.
With respect to claim(s) 7 and 15, the claim(s) recite(s) attempting to compile, which reads on a human ensuring the code is in a machine readable format and checking to make sure it can be compiled. No additional limitations are present.
With respect to claim(s) 8 and 16, the claim(s) recite(s) performing one or more tests, which reads on a human following the code as written and using previously determined additional information, to see if the code works as intended. No additional limitations are present.
These claims further do not remedy the judicial exception being integrated into a practical application and further fail to include additional elements that are sufficient to amount to significantly more than the judicial exception.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1, 3-7, 9, 11-15, 17, 19, and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Jiang et al. (‘Discovering the Syntax and Strategies of Natural Language Programming with Generative Language Models’, ACM, 2022), hereinafter Jiang, in view of Wang et al. (U.S. PG Pub No. 2022/0019932), as found in the IDS, hereinafter Wang.
Regarding claims 1, 9, and 17, Jiang teaches
(claim 1) A system comprising (a code synthesis tool enabling user input Fig.1,(Abstract, Intro)):
(claim 1) a generative artificial intelligence (GAI) model (natural language code synthesis tool using a large generative language model, i.e. generative artificial intelligence (GAI) model (Abstract));
(claim 1) a framework server configured to perform operations comprising (human computer interactions and computing, i.e. framework server configured to perform Fig. 1,(CCS concepts, Abstract, Intro)):
(claim 9) A method comprising (a code synthesis tool enabling user input Fig.1,(Abstract, Intro)):
(claim 17) …instructions which, when executed by one or more processors, cause the one or more processors to perform operations comprising (human computer interactions and computing, i.e. one or more processors…to perform operations, where the natural language code synthesis tool includes models and task-specific prompts, i.e. instructions, Fig. 1,(CCS concepts, Abstract, Intro, Sec. 3.1 and 3.2)):
receiving, from a framework client, a text-based prompt describing one or more services (the user can input into an interface, i.e. from a framework client, a text input for a prompt in a natural language format, i.e. receiving…a text-based prompt describing, as a description of a request for code, such as making a button or creating a flashcard app, i.e. describing one or more services Fig. 1,(Abstract, Intro));
creating a server generation request comprising the text-based prompt and a first pre-designed system message (when a tag is detected related to the input, i.e. comprising the text-based prompt, the input is incorporated into a task-specific prompt with examples is automatically loaded, i.e. a first pre-designed system message, and the prompt is loaded as input into the generative language model to execute the task, such as generating code, i.e. creating a server generation request Fig. 1,(Intro, Sec. 3.1 and 3.2, Appendix A and B));
feeding the server generation request to the GAI model to generate one or more server files specifying a declarative-based data model definition for a software-based server service, the one or more server files being …executable in the framework server (the task-specific prompt with incorporated user input, i.e. server generation request, fed into the generative language model, i.e. feeding…to the GAI model, to output code such as HTML or JavaScript and/or subtask suggestions, i.e. generate one or more server files specifying a declarative-based data model definition, for a request such as making a To-Do app or a flashcard app, where the code result can be demonstrated, such as displaying the “Submit” button coded for in HTML, i.e. a software-based server service the one or more server files being…executable in the framework server Fig. 1,(Intro, Sec. 3.1 and 3.2, Appendix A and B)).
While Jiang provides generating code for apps from natural language using an LLM, Jiang does not specifically teach storing the files in a repository or compiling files, and thus does not teach
a server repository; and
the one or more server files being compilable; and
storing the one or more server files in the server repository.
Wang, however, teaches a server repository (the server system includes a data store [0030]); and
the one or more server files being compilable (the code project, such as a Java project, is compiled and packaged, i.e. one or more server files being compilable [0057]); and
storing the one or more server files in the server repository (generated files are written into the data store [0038],[0052-5]).
Wang also specifically teaches A non-transitory machine-readable medium storing instructions…([0079-80])
Jiang and Wang are analogous art because they are from a similar field of endeavor in developing service code in response to user input. Thus, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the generating code for apps from natural language using an LLM teachings of Jiang with the compilation of code and storage of generated files for use the by the system as taught by Wang. It would have been obvious to combine the references to automatically generate services from user text and image input using machine learning (Wang [0004-5]).
Regarding claims 3, 11, and 19, Jiang in view of Wang teaches claims 1, 9, and 17, and Wang further teaches
the one or more server files include a schema file and a service file (an EDM file is generated based on the user input and is provided in a Common Schema Definition Language, i.e. include a schema file, which is further used to generate service code, i.e. a service file [0026-7],[0055]).
Where Jiang specifically teaches that different kinds of outputs are generated by the LLM (Intro, Sec. 3.1, Sec. B intro and B.1).
And where the motivation to combine is the same as previously presented.
Regarding claims 4, 12, and 20, Jiang in view of Wang teaches claims 3, 11, and 19, and Wang further teaches
the schema file and the service file define an Open Data Protocol (OData) service (an EDM file is generated based on the user input and is provided in a Common Schema Definition Language, i.e. include a schema file, which is further used to generate OData service code, i.e. a service file define an OData service [0026-7],[0055]).
Where the motivation to combine is the same as previously presented.
Regarding claims 5 and 13, Jiang in view of Wang teaches claims 1 and 13, and Jiang further teaches
creating a data generation request comprising one or more statements generated from the one or more server files and a second pre-designed system message (the user can input a text input for a prompt in a natural language format, such as a step from a suggestion for how to create an app from suggestions provided in response to a previous prompt, as a description of a request for code, i.e. one or more statements generated from the one or more server files, and when a tag is detected related to the input, the input is incorporated into a task-specific prompt with examples is automatically loaded, such how to fix the code, output code in a different language, or provide further suggestions for sub-tasks, i.e. a second pre-designed system message, and the prompt is loaded as input into the generative language model to execute the task, i.e. creating a data generation request Fig. 1,(Intro, Sec. 3.1 and 3.2, Appendix A and B));
feeding the data generation request to the GAI model to generate one or more server data files (the task-specific prompt with incorporated user input, i.e. server generation request, fed into the generative language model, i.e. feeding…to the GAI model, to output code and/or subtask suggestions, i.e. generate one or more server data files, for a request such as making a To-Do app or a flashcard app Fig. 1,(Intro, Sec. 3.1 and 3.2, Appendix A and B)).
Where Wang further teaches storing the one or more server data files in a data repository (generated files are written into the data store, where demo data can be generated and further populated into the Java project, i.e. one or more server data files [0038],[0052-5],[0059]).
And where the motivation to combine is the same as previously presented.
Regarding claims 6 and 14, Jiang in view of Wang teaches claims 5 and 13, and Jiang further teaches
validating the one or more server files (for prompts that produce HTML, i.e. one or more server files, GenLine renders the model output in an HTML iframe, providing a way to validate the output at a glance, i.e. validating (Sec. 3.1)); and
in response to a determination that the one or more server files have failed the validation, retrying the feeding of the server generation request to the GAI model (if there are errors in the code noticed by validating the output at a glance, i.e. in response to a determination that the one or more server files have failed the validation, a prompt template for fixing errors in existing code can be sent to the LLM that incorporates user input, which can include code and natural language, i.e. retrying the feeding of the server generation request to the GAI model Fig. 1,(Intro, Sec. 3.1 and 3.2, Appendix A and B)).
Regarding claims 7 and 15, Jiang in view of Wang teaches claims 6 and 15, and Jiang further teaches
the validating includes attempting to compile the one or more server files (for prompts that produce HTML, i.e. one or more server files, GenLine renders the model output in an HTML iframe, providing a way to validate the output at a glance, i.e. validating includes attempting to compile Fig. 1,(Sec. 3.1)).
Claim(s) 8 and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Jiang, in view of Wang, and further in view of Kulal et al. (‘SPoC: Search-based Pseudocode to Code’, NeurIPS 2019), hereinafter Kulal.
Regarding claims 8 and 16, Jiang in view of Wang teaches claims 6 and 14.
While Jiang in view of Wang provides validating the files, Jiang in view of Wang does not specifically teach performing tests, and thus does not teach
the validating includes performing one or more tests on the one or more server files using the one or more server data files.
Kulal, however, teaches the validating includes performing one or more tests on the one or more server files using the one or more server data files (a synthesized program, i.e. one or more server files, is accepted, i.e. validating, if it successfully compiles and passes all public test cases, i.e. performing on or more tests, where the text cases are input-output sets that the program must compute correctly, i.e. one or more server data files Figs. 1 and 2,(Intro, Sec. 2)).
Where Wang teaches demo data that is generated and populated into a Java project [0055].
Jiang, Wang, and Kulal are analogous art because they are from a similar field of endeavor in developing service code in response to user input. Thus, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the validating the files teachings of Jiang, as modified by Wang, with performing tests with text cases for validation as taught by Kulal. It would have been obvious to combine the references to improve the program synthesis success rate during program validation (Kulal Abstract).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NICOLE A K SCHMIEDER whose telephone number is (571)270-1474. The examiner can normally be reached 8:00 - 5:00 M-F.
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, Pierre-Louis Desir can be reached at (571) 272-7799. 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.
/NICOLE A K SCHMIEDER/Primary Examiner, Art Unit 2659
Prosecution Insights