DETAILED ACTION
This action is in response to the initial filing of application no. 18/436,673 on 02/08/2024.
Claims 1 – 20 are still pending in this application, with claims 1, 10 and 13 being independent.
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 .
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1 – 5, 7, 10 – 17 and 19 is/are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Baruch et al. (US 2024/0354503) (“Baruch”).
As to claims 1 and 13, Baruch discloses a system (Abstract) comprising one or more processors (Fig.14, 1402; [0304] [0307 - 0309]) and memory (Fig.14, 1404) storing instructions (Fig.14, 1414) that, in response to execution by the one or more processors ([0316]), cause the one or more processors to: processing a large language model (LLM) ([0052 – 0054] [0065] [0067] [0068] [0137 – 0140]) prompt using an LLM (Fig.1, 124 and Fig.2, 222; [0051] [0069] [0071] [0100]) to generate a first LLM response (draft post) (Fig.1, 126 and Fig.3D, 342; [0074] [0075] [0140]); providing the first LLM response to a client application (content generation assistant, Fig.1, 128; [0051]) ([0075] [0140]), wherein the first LLM response is operable by the client application to cause first content derived from the first LLM response to be rendered at one or more output devices (Fig.3D, 342; [0075] [0140]); assembling, as a second LLM prompt, at least a portion of the first LLM response (The portion comprises a subpart of the draft post selected by user) with data indicative of a request to generate one or more recommended alterations to the portion of the first LLM response (A prompt(s) comprising an instruction to rewrite the subpart of the first LLM response selected by a user with more descriptive detail is generated and provided to the generative language model, Fig.3E, 356, Fig.3F, 364 and Fig.3G, 372; [0141 – 0144]); processing the second LLM prompt using the LLM to generate a second LLM response ([0144]); and providing at least a portion of the second LLM response to the client application (Fig.3G, 374; [0075] [0144]), wherein the portion of the second LLM response is operable by the client application to cause one or more selectable elements (Fig.3G, 376 and 378) to be rendered at one or more of the output devices ([0075] [0144]), wherein each of the one or more selectable elements: identifies a respective one of the recommended alterations (The accept mechanism identifies a first, recommended descriptive detail/alteration of the selected portion of the first response; and the try another mechanism identifies both none and a subsequent, recommended descriptive detail/alteration of the selected portion of the first response., [0144]), and is operable to cause corresponding modified content to be rendered at one or more of the output devices (The user selects the accept and post mechanism(s) which causes the application software to initiate distribution of content item containing the content output by the generative language model to one or more other users of the application software system or user network, Fig.3G, 378 and Fig.3H, 386; [0075] [0081] [0082] [0141] [0144] [0145]), wherein the modified content includes at least some of the first content modified based on the respective one of the recommended alterations ([0075] [0081] [0082] [0141] [0144] [0145]).
As to claims 2 and 14, Baruch further teaches: receiving, from the client application, an indication that one or more of the selectable elements was operated (The “try another” mechanism is selected., [0144]); and in response to the receiving, providing a remainder of the second LLM response to the client application (The second LLM prompt comprises both the initial and modified prompt to rewrite the subpart with more descriptive detail. The remainder of the second LLM response comprises the regenerated, suggested revision to the portion/subpart. The remainder is provided to the user within the thought starter window displayed on the user interface., [0075] [0144] ]), wherein the modified content corresponding to the operated selectable element is derived from the remainder of the second LLM response (The user selects the accept mechanism to accept the regenerated, suggested revision. The accept mechanism causes the suggested revision/modified content to be incorporated in to the text displayed on the user device via a content generation assistant, Fig.3G, 378 and Fig.3H, 386; [0075] [0081] [0082] [0141] [0144] [0145])
As to claims 3 and 15, Baruch further teaches: receiving, from the client application, an indication that one or more of the selectable elements was operated (The tone of voice option in the writing assistant menu is selected, Fig.3I, 392; [0079] [0146]); in response to the receiving: assembling, as a third LLM prompt, at least a portion of the first LLM response and data indicative of the recommended alteration identified by the operated selectable element (A third prompt is generated to rewrite the draft post using a different tone. The draft post comprises a portion of the first LLM response and the altered subpart of the draft post, wherein the altered subpart comprises the content which was written with more descriptive detail., Fig.3G, 370, 374, 378, Fig.3H, 384 and Fig.3I, 390 and 392; [0035] [0103] [0146]); processing the third LLM prompt using the LLM to generate a third LLM response ([0146]), and providing at least a portion of the third LLM response to the client application (A response generated by the generative AI model is output to the client application, [0075] [0146] [0293] [0295] [0297]), wherein the portion of the third LLM response is operable by the client application to cause the modified content to be rendered at one or more of the output devices (The user selects the post mechanism(s) which causes the application software to initiate distribution of content item containing the content output by the generative language model to one or more other users of the application software system or user network, Fig.3I, Post; [0075] [0141] [0146]).
As to claims 4 and 16, Baruch further teaches; receiving, from the client application, an indication of a subportion of the first content that has been selected using one or more input devices (Fig.1, 129 and Fig.3G, 372; [0079] [0080] [0144]); and extracting, as the portion of the first LLM response that is assembled into the second LLM prompt, a subportion of the first LLM response that corresponds to the selected subportion of the first content ([0144]).
As to claims 5 and 17, Baruch further teaches, wherein the portion of the second LLM response provided to the client application includes the entire second LLM response (Fig.3G, 374 and Fig.3H, 384; [0144] [0145]).
As to claims 7 and 19, Baruch further teaches, wherein a given selectable element of the one or more selectable elements comprises a textual summary (“Try adding more descriptive detail”) of the modified content that is rendered in response to operation of the given selectable element (Fig.3G, 374 and 378; [0144]).
As to claim 10, Baruch teaches a method (Abstract) implemented using one or more processors (Fig.14, 1402; [0304] [0307 - 0309]), comprising: receiving, from a user (Fig.1, 129; [0079]) via a client application (content generation assistant, Fig.1, 128; [0051]), a user-composed content (Fig.3A, 302, Fig.3B, 324 and 326, Fig.3C, 334; [0134 – 0139]); assembling, as a first large language model (LLM) (Fig.1, 124 and Fig.2, 222; [0051] [0069] [0071] [0100]) prompt, data indicative of the user-composed content (The data includes a subpart of the draft post generated based on user input.) and data indicative of a request to generate one or more recommended alterations to the user-composed content (The data includes instructions to rewrite the subpart with more descriptive detail) (Fig.3E, 356, Fig.3F, 364 and Fig.3G, 372; [0140- 0144]); processing the first LLM prompt using an LLM to generate a first LLM response (Fig.3G, 374; [0144]); and providing at least a portion of the first LLM response to the client application ([0075] [0144] [0293] [0295] [0297]), wherein the portion of the first LLM response is operable by the client application to cause one or more selectable elements (Fig.3G, 376 and 378, Try another and Accept) to be rendered at one or more output devices (Fig.3G,374; [0075] [0144]), wherein each of the one or more selectable elements: identifies a respective one of the recommended alterations (The accept mechanism identifies a first, recommended descriptive detail/alteration of the selected portion of the first response; and the try another mechanism identifies a subsequent, recommended descriptive detail/alteration of the selected portion of the first response., [0144]), and is operable to cause corresponding modified content to be rendered at one or more of the output devices (The user selects the accept and post mechanism(s) which causes the application software to initiate distribution of content item containing the content output by the generative language model to one or more other users of the application software system or user network, Fig.3G; [0075] [0144]), wherein the modified content includes at least some of the user-composed content modified based on the respective one of the recommended alterations (Fig.3G, 372 and 374; [0144]).
As to claim 11, Baruch further teaches: receiving, from the client application, an indication that one or more of the selectable elements was operated (The try another mechanism is selected., Fig.3G, 376; [0144]); and in response to the receiving, providing a remainder of the first LLM response to the client application (The first LLM The remainder of the first LLM response comprises the regenerated, suggested revision to the portion/subpart. The remainder is provided to the user within the thought starter window displayed on the user interface., [0144]), wherein the modified content corresponding to the operated selectable element is derived from the remainder of the second LLM response (The user selects the accept mechanism to accept the regenerated, suggested revision., Fig.3G, 378 and Fig.3H; [0144] [0145]).
As to claim 12, Baruch further teaches, receiving, from the client application, an indication that one or more of the selectable elements was operated (The try another mechanism is selected., Fig.3G, 376; [0144]); in response to the receiving: assembling, as a second LLM prompt, at least a portion of the user-composed content and data indicative of the recommended alteration identified by the operated selectable element (A new prompt is generated with the selected subpart and instructions to rewrite the subpart, wherein the instructions indicate the recommended alterations identified by the operated selected element, “try another.” The recommended alteration identified by “try another” is broadly, reasonably interpreted as none., [0144], processing the second LLM prompt using the LLM to generate a second LLM response ([0144]), and providing at least a portion of the second LLM response to the client application Fig.3G, 374; [0075] [0144]), wherein the portion of the second LLM response is operable by the client application to cause the modified content to be rendered at one or more of the output devices (The user selects the post mechanism(s) which causes the application software to initiate distribution of content item containing the content output by the generative language model to one or more other users of the application software system or user network, Fig.3I, Post; [0075] [0141] [0146]).
Claim Rejections - 35 USC § 103
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) 6 and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Baruch et al. (US 2024/0354503) (“Baruch”) in view of Maschmeyer et al. (US 2024/0320444) (“Maschmeyer”).
For claim 6 and 18, Baruch fails to teach wherein at least the portion of the second LLM response is operable by the client application to locally cache one or more instances of modified content corresponding to one or more of the selectable elements.
However, Maschmeyer discloses a method for AI-guided content generation (Abstract), comprising the following: a response generated by a LLM (generative AI model, Fig.1, 112; [0030] [0077] [0082]) is operable by a client application (content generation engine comprising a user interface module, Fig.1, 114 and 116, Fig.5A, 500 [0123]); [0077] [0080] [0083 – 0085]) to locally store one or more instances of content (Fig.5A, 509a) corresponding to one or more selectable element (Keep, Fig.5A, 506b) (“For example, the user can select/highlight text within the first text output 510a. The selected text 509a can be stored for inclusion in a final text output by selecting the “Keep” button 506b. The stored data for preferred text portions may include the actual selected text, relative location of the selection, edits to the selected text (if any), and the like.”, [0124] [0125]).
Therefore, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve Baruch’s invention in the same way that Maschmeyer’s invention has been improved to achieve the following, predictable results for the purpose of efficiently producing distributable content using generative models which reduce the amount of resources and time used to generate the distributable content (Baruch, [0021 – 0023]): at least the portion of the second LLM response is further operable by the client application to locally cache (store) one or more instances of content generated by the LLM, e.g. modified content (corresponding to one or more of the selectable elements.
Claim(s) 8 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Baruch et al. (US 2024/0354503) (“Baruch”) in view of Woulfe et al. (US 2025/0209389) (“Woulfe”) and further in view of Joshi et al. (“With Great Power Comes Great Responsibility!": Student and Instructor Perspectives on the influence of LLMs on Undergraduate Engineering Education”) (“Joshi”).
For claims 8 and 20, Baruch fails to teach the following: wherein the first LLM response comprises one or more time descriptions forming an incomplete timeline, and one or more of the recommended alterations includes an additional time description to fill a gap in the incomplete timeline.
However, Woulfe discloses a data processing system (Abstract), comprising the following: a response from an LLM comprises one or more time descriptions forming an incomplete timeline (schedule generated by an AI-assisted schedule planning application, wherein the schedule is missing an event, e.g. time for breakfast, [0015 – 0018] [0052] [0062 – 0067]); and a recommended alteration (field in the user interface to add an event to a timeline) includes an additional time description (“add time for breakfast”) to fill a gap in the incomplete timeline ([0067] [0068]).
Moreover, Joshi discloses a system and method for utilizing a LLM (Abstract), wherein a LLM (e.g. ChatGPT) is used as a writing assistant (“Many students suggested that ChatGPT is capable of writing high quality theoretical and verbose content, and hence is useful for assignments that require polished, effective use of language. Generating descriptive essays, creative writing tasks, and reports are some popular language-related uses of ChatGPT among students.”, 4.2 Student Perspective on ChatGPT - Qualitative Evaluation, 4.2.1 Transitioning Workflows - The Advantages of Leveraging ChatGPT) and schedule generation (“Participants have also used ChatGPT to manage schedules, create timetables.”, 4.2 Student Perspective on ChatGPT - Qualitative Evaluation, 4.2.1 Transitioning Workflows - The Advantages of Leveraging ChatGPT).
Therefore, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve Baruch’s invention in the same way that Woulfe’s invention has been improved to achieve the following, predictable results for the purpose of providing an improved user experience by leveraging a LLM to further create and modify a schedule consisting of tasks assigned to discrete time slots with inferred actions (Woulfe, [0016] [0017]): a LLM response further comprises one or more time descriptions forming an incomplete timeline, and one or more recommended alterations includes an additional time description to fill a gap in the incomplete timeline.
Additionally, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to modify the invention disclosed by the combination of Baruch and Woulfe in the same way that Joshi’s invention has been improved to achieve the following, predictable results for the purpose of enhancing user experience by enabling a user to interact with a multi-task LLM: the system further comprises user interfaces for a user to engage with the system to receive writing and scheduling assistance; the first LLM response further comprises one or more time descriptions forming an incomplete timeline, and the one or more of the recommended alterations further includes an additional time description to fill a gap in the incomplete timeline.
Claim(s) 9 is/are rejected under 35 U.S.C. 103 as being unpatentable over Baruch et al. (US 2024/0354503) (“Baruch”) in view of Novak et al. (US 2018/0373724) (“Novak”).
For claim 9, Baruch fails to teach the following: formulating a search query based on one or more details of the first LLM response; and retrieving, from a search engine, one or more documents that are responsive to the search query; wherein a given selectable element of the one or more selectable elements is operable to cause additional content from the one or more documents that are responsive to the search query to be included in the second LLM prompt.
However, Baruch further discloses that one or more selectable elements include an element to attach a file to the generated text (Fig.3I. 392; [0146]). Furthermore, an accepted attachment is included in the second LLM prompt when the generated text including the accepted attachment is provided to the LLM to be rewritten ([0144]).
Additionally, Novak discloses a system and method for attaching an inline content file (Abstract), comprising the following: formulating a search query based on text generated by a content authoring application (Fig.3A; [0035] [0036]); generating a search query based on or more details of the text (Fig.3A, 126, 128, 204; [0035 – 0037]); retrieving, from a search engine (file searcher, Fig.1, 122; [0026]), one or more documents that are responsive to the search query (Fig.3B, 306, 308, 310, 312; [0037]); and generating one or more selectable elements (Fig.3B, 306, 308, 310, 312), wherein the one or more selectable elements are operable to cause additional content from the one or more documents that are responsive to the search query to be attached to the text (Fig.3C, 206 and 208a, Fig.3D, 204 and 212; [0038]).
Therefore, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve Baruch’s invention in the same way that Novak’s invention has been improved to achieve the following, predictable results for the for the purpose of efficiently producing distributable content using generative models which reduce the amount of resources and time used to generate the distributable content (Baruch, [0021 – 0023]): the system further comprises a search engine; a search query is further formulated based on one or more details of the first LLM response based on a request to add an attachment; one or more documents that are responsive to the search query are retrieved from the search engine; the selectable elements further comprise one or more selectable elements which are operable to cause additional content from the one or more documents that are responsive to the search query to be providing in the text, wherein the text is included in the second LLM prompt when the text is selected to be rewritten.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SONIA L GAY whose telephone number is (571)270-1951. The examiner can normally be reached Monday-Friday 9-5 ET.
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, Daniel Washburn can be reached at 571-272-5551. 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.
/SONIA L GAY/Primary Examiner, Art Unit 2657