DETAILED ACTION
The following FINAL Office Action is in response to Applicant’s Response filed on 12/04/2025.
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 . 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 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.
Status of Claims
Claims 1-18 were previously pending and subject to a non-final Office Action mailed 09/12/2025. Claims 1, 2, 6, 8, and 18 were amended. Claims 1-18 are currently pending and are subject to the final Office Action below.
Response to Arguments
35 USC § 112
Regarding the Claims 8-11 rejection, Applicant has amended Claim 8 as Examiner suggested. Accordingly, the rejection is withdrawn.
Regarding the Claims 2-12 rejection, Applicant has amended Claim 2 by removing “useful”. However, Applicant has not amended Claim 4 and 12. Accordingly, the rejection is maintained in part.
Regarding the Claim 6 rejection, Applicant has amended Claim 6 by removing “appropriate”. Accordingly, the rejection is withdrawn.
Regarding the Claim 16 rejection, Applicant has amended Claim 16 by removing “intelligently”. Accordingly, the rejection is withdrawn.
35 USC § 101
Applicant’s arguments, see page 7, filed 12/04/2025, with respect to the 35 U.S.C. 101 rejections of Claims 1-18 have been fully considered and are not persuasive.
Applicant argues that claim 1 is directed to patentable subject matter. Examiner respectfully disagrees. Generating a new workflow where the workflow includes programming code for performing the new service is part of the abstract idea of certain methods of organizing human activity.
As described in specification paragraph 23, “Even the most advanced cutting-edge workflow automation systems are reliant only on the enterprise side to create a workflow. This disclosure truly shifts the paradigm by allowing the end-users to participate meaningfully in workflow automation within the enterprise, that too without having to code. This dramatic paradigm shift enables end-users to receive a truly personalized service from the enterprise, and the paradigm shift is made possible by Generative Artificial Intelligence (Generative AI).”
Thus, the enterprise persona generates the workflow for the user (for example, an employee) as a service and then, the workflow may be stored as a template for use by other employees as described in specification paragraph 45-48, the user provides input to the enterprise persona and in paragraph 49, the enterprise persona sets boundaries on the kinds of services the user can generate and if approved, the workflows become part of the enterprise service and are available to other users.
Accordingly, the 35 U.S.C. 101 rejection of Claims 1-18 is maintained.
35 USC § 103
Applicant’s arguments, see pages 7-11, filed 12/04/2025, with respect to the 35 U.S.C. 103 rejections of Claims 1-18 have been fully considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Examiner relies on Betteridge and new reference Sullivan to teach the bolded feature of Claim 1 that Applicant noted.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 4-12 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 4 recites “responsive to determining that the new workflow is associated with a useful service” and Claim 12 recites “responsive to determining that the new workflow is not associated with a useful service”.
The term “useful” recited in Claims 4 and 12 is a subjective term which renders the claim indefinite. Applicant’s specification paragraph 49 states “If the generated workflow is deemed by the enterprise persona to be as an useful service at step 320, then the workflow is approved (330). The approved workflows are now part of the enterprise service and are available for the other end-users as well. If the generated workflow is deemed by the enterprise persona not to be as an useful service at step 320, then the workflow is rejected (340).”
The specification does not provide an objective standard for measuring the scope of the term “useful”, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. Dependent Claims 5-11 inherit the rejection as they do not cure the deficiencies of claim 4.
For purposes of the art rejection, Examiner interprets useful as accurate as in determining whether the new workflow is accurate and thus, would be useful to end-users of the enterprise.
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 1-18 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1
Claims 1-18 are directed to a method (i.e., a process). Therefore, the claims all fall within one of the four statutory categories of invention.
Step 2A Prong 1
Independent Claim 1 recites:
collecting the end-user’s input during runtime of a selected pre-created workflow module that the end-user is currently interacting with;
analyzing the end-user’s input…to determine the end-user’s expressed intent; and
responsive to determining, that the end-user’s expressed intent is not covered by any of the plurality of pre-created workflow modules, generating a new workflow corresponding to a new service that is associated with the end-user’s expressed intent, wherein the new workflow comprises programming code for performing the new service
The limitations stated above are processes that under broadest reasonable interpretation covers “certain methods of organizing human activity” (“commercial interactions” or “managing personal behavior or relations or interactions between people”). Specifically, interactions between an end-user and an enterprise persona in light of Applicant’s specification paragraph 5 “With the rise of the Large Language Models (LLM) and the generative capability of an AI-powered platform, the existing micro-engagement capabilities can be enhanced in a consumer-friendly way, if an appropriate set of questions are asked, adequate responses are provided to advance interaction, and/or proper interfaces are provided to the end-user to express their needs. End-user's interest in providing information depends a lot on the enterprise persona with whom the end-user is engaging. Therefore, the enterprise persona needs to be empowered with tools to generate fitting engagement texts.”
Accordingly, the claim recites an abstract idea.
Step 2A Prong 2
This judicial exception is not integrated into a practical application. The independent claim also recites “providing, by an enterprise persona, using a micro-engagement engine, a visual interface for the end-user to access and interact with a plurality of pre-created workflow modules, each of the plurality of pre-created workflow modules corresponding to a respective service that the enterprise is currently capable of providing to the end-user” and “a backend of the micro-engagement engine”.
Thus, Claim 1 recites the additional elements of an enterprise persona, a micro-engagement engine, a visual interface, a backend of the micro-engagement engine which are all recited at a high-level of generality (generic computer/functions) such that when viewed as a whole/ordered combination, it amounts to no more than mere instructions to apply the judicial exception using generic computer components. See MPEP 2106.05(f).
Further, “providing, by an enterprise persona, using a micro-engagement engine, a visual interface for the end-user to access and interact with a plurality of pre-created workflow modules, each of the plurality of pre-created workflow modules corresponding to a respective service that the enterprise is currently capable of providing to the end-user” limits the abstract idea to a particular field of use by specifying that the abstract idea relates to activities that are executed in visual interface environment (provided by an enterprise persona using the micro-engagement engine). The visual interface environment allowing end-users to access and interact with the pre-created workflow modules.
Thus, the claim as a whole, looking at additional elements individually and in combination, does not integrate the judicial exception into a practical application as the additional elements are mere instructions to apply the judicial exception using generic computer components and field of use which does not impose meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
Step 2B
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of an enterprise persona, a micro-engagement engine, a visual interface, a backend of the micro-engagement engine to perform the steps/functions recited above amounts to no more than mere instructions to apply the exception using a generic computer. Mere instructions to apply the exception using a generic computer component cannot provide an inventive concept.
Again, “providing, by an enterprise persona, using a micro-engagement engine, a visual interface for the end-user to access and interact with a plurality of pre-created workflow modules, each of the plurality of pre-created workflow modules corresponding to a respective service that the enterprise is currently capable of providing to the end-user” limits the abstract idea to a particular field of use by specifying that the abstract idea relates to activities that are executed in visual interface environment (provided by an enterprise persona using the micro-engagement engine).
None of the steps of Claim 1 when evaluated individually or as an ordered combination amount to significantly more than the abstract idea. The additional elements are merely used to perform the limitations directed to the abstract idea and amount to no more than mere instructions to apply the exception using a generic computer or field of use, thus, the analysis does not change when considered as an ordered combination. Further, the additional elements do not meaningfully limit the claim. Accordingly, Claim 1 is ineligible.
Dependent Claims 2, 4, and 12 further specify determining whether the new workflow is associated with a [useful] service and storing or rejecting the new workflow accordingly.
Dependent Claim 3 further specifies wherein the enterprise persona is a real person or a virtual person which is further narrowing the abstract idea identified above. Regarding the virtual person which is recited at a high-level of generality (generic computer/functions) such that when viewed as a whole/ordered combination, it amounts to no more than mere instructions to apply the judicial exception using generic computer components.
Dependent Claims 5-6 add an additional element of a scheduler module which decides a time to deploy the new workflow. The scheduler module is recited at a high-level of generality (generic computer/functions) such that when viewed as a whole/ordered combination, it amounts to no more than mere instructions to apply the judicial exception using generic computer components.
Dependent Claims 7-11 add an additional element of large language model (LLM) which may be trained, fine-tuned, receive inputs, and generate a JSON representation. The additional element generally links the judicial exception to a particular technological environment, particularly, for use with LLMs as the new workflow determined as part of the abstract idea is used to train the LLM.
Dependent Claim 13 adds an additional element of a conversational agent to collect the end-user’s input and Claim 14 adds an additional element of a language intelligence services architecture or LISA which analyzes the input. The conversational agent and LISA are recited at a high-level of generality (generic computer/functions) such that when viewed as a whole/ordered combination, it amounts to no more than mere instructions to apply the judicial exception using generic computer components.
Dependent Claim 15 adds an additional element of a Document Intelligence Services Architecture (DISA) to analyze the input and Claim 16 adds an additional element of a generative flow builder module to create prompts to collect as input. Both elements recited at a high-level of generality (generic computer/functions) such that when viewed as a whole/ordered combination, it amounts to no more than mere instructions to apply the judicial exception using generic computer components.
Dependent Claims 17-18 further specify the interaction with the user occurs via a plurality of communication channels and what the channels may be.
Such limitations are further directed towards organizing human activity between an end-user and an enterprise persona and the additional elements are mere instructions to apply the judicial exception or field of use. Thus, taken alone and when viewed in combination, nothing in dependent claims 2-18 adds additional elements that are sufficient to amount to significantly more than the judicial exception. Claims 1-18 are ineligible.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 13-14, and 16-18 are rejected under 35 U.S.C. 103 as being unpatentable over Betteridge et al. (US2024/0104469) in view of Sullivan et al. (US 10,554,817).
As per independent Claim 1,
Betteridge teaches a method for integrating workflows dynamically generated by an end-user during an automated interactive engagement session between the end-user and an enterprise, the method comprising:
providing, by an enterprise persona, using a micro-engagement engine, a visual interface for the end-user to access and interact with a plurality of pre-created workflow modules each of the plurality of pre-created workflow modules corresponding to a respective service that the enterprise is currently capable of providing to the end-user (para. 26-27 software servers of a UCaaS platform (“micro-engagement engine”) deliverable to a client; see para. 13 where UCaaS platforms implement services usable for contact center engagements which are used to provide support actions in response to requests for technical support or complaints, the support actions can be provided using workflows (“pre-created workflow modules”); figure 3 and para. 45-47 software platform; figure 4 and para. 56-59 where the user engages with a call center or contact center service and the agent (“enterprise persona”) may be a human or non-human; para. 60-61 graphical user interface on the user device (“visual interface”) is used to display interactions between the user and the agent; para. 62 user transmits a query to obtain a remedy for a defective product; see also para. 19 “during execution of a workflow”, para. 98 “query may be received during an interaction between the user device and an agent device”, and para. 67 workflow database (“pre-created workflow modules”))
collecting the end-user’s input during runtime of a selected pre-created workflow module that the end-user is currently interacting with (para. 59 communications between the user and agent are monitored; para. 62 user transmits a query and the user may provide evaluation responses; para. 65 and figure 5 user device transmits a query to initiate a contact center engagement)
analyzing the end-user’s input at a backend of the micro-engagement engine to determine the end-user’s expressed intent (para. 60 evaluation integration software analyzes the text to determine the intent of a query received from the user device; para. 66 determining the query intent by analyzing for keywords)
responsive to determining, that the end-user’s expressed intent is not covered by any of the plurality of pre-created workflow modules, generating a new action (para. 15 integrating evaluations into an executed workflow that allows an entity to customize the text copy, responses, and custom action types for an evaluation; figure 5 para. 67-68 where the method determines whether a workflow for the determined intent exists and para. 69 if it is determined that a workflow for the determined intent does not exist, a determination of whether a support action exists is made and if a support action exists, it is transmitted to the user device)
Betteridge does not teach, but Sullivan teaches:
Generating a new workflow corresponding to a new service that is associated with the end-user’s expressed intent, wherein the new workflow comprises programming code for performing the new service (Col. 23 Line 64-Col. 24 Line 16 newly discovered intent (“end-user’s expressed intent”), Col. 24 Line 17-41 system identifies available actions for the intent where the actions may have been previously unknown to the system, Col. 24 Lines 42-57 system designs one or more contact workflows for the intent (“new workflow”) and configure the automated service agent to detect the intent and access each API to perform the actions (“programming code”); Col 8 Lines 20-33 record for intent may store information describing the action(s) the service agent can take to satisfy the intent where the action can include one or more computer functions to execute such as APIs for obtaining or modifying data; Col. 12 Line 63 to Col. 13 Line 45 where in the first paragraph, the automated service agents or chatbots execute the APIs and in the second paragraph, the automated service agent may be associated with a discovered intent; see also Col. 36 Line 5-21 “generate an automated service agent data object including program code for executing the contact workflow”)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Sullivan with the motivation of improving the performance of the agent/enterprise persona and reduce the amount of work a human would have to perform.
See Col. 4 Lines 21-44 “reduce the amount of collaborative activity that a particular user must manually perform” and Col. 5 Lines 35-40 “to service a customer contact without human intervention and/or to reduce the amount and/or improve the accuracy of actions a human service agent would otherwise perform manually to execute the contact workflow”.
As per dependent Claim 13,
Betteridge/Sullivan teaches the method of claim 1.
Betteridge teaches:
wherein the micro-engagement engine uses a conversational agent to collect the end-user’s input (para. 58 contact center agent; para. 61-62 interaction between the user and agent such as a conversation; see also para. 57 and para. 70)
As per dependent Claim 14,
Betteridge/Sullivan teaches the method of claim 13.
Betteridge teaches:
wherein a software analyzes the end-user’s input collected by the conversational agent (para. 60 analyzes the text for keywords to determine the intent of the query received from the user device and the context of the keywords may be determined using a machine learning model)
Betteridge does not explicitly teach, but Sullivan teaches:
wherein a Language Intelligence Services Architecture (LISA) analyzes the end-user’s input collected by the conversational agent (Col. 24 Line 58 – Col. 25 Line 19 natural language processing system to analyze the contact between a customer and an agent; see also Col. 25 Line 41-61, Col. 23 Line 18-63, and Col. 30 Line 50-53)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Sullivan since each individual element and its function are shown in the prior art, albeit shown in separate references, the difference between the claimed subject matter and the prior art rests not on any individual element or function but in the very combination itself. That is in the substitution of LISA of Sullivan for the software of Betteridge. Both are elements or techniques utilized for intent analysis of the user’s input. Thus, the simple substitution of one known element for another producing a predictable result renders the claim obvious.
As per dependent Claim 16,
Betteridge/Sullivan teaches the method of claim 1.
Betteridge teaches:
wherein a generative flow builder module creates prompts to collect a series of responses from the end-user as the end-user’s input (para. 16 display a support action and request input from the user as to whether the support action was helpful and if helpful, the system will trigger the integrated evaluation; para. 69-70 evaluation questions enable the system to obtain a comment from the user; para. 71-74 and 75-85 where the system solicits ratings and comments from the user)
As per dependent Claim 17,
Betteridge/Sullivan teaches the method of claim 1.
Betteridge further teaches:
wherein the micro-engagement engine supports interaction with the end-user via a plurality of communication channels (para. 57 user and agent device communicate with each other using communications that include SMS, email, webchat, VOIP (voice over internet protocol), video, social media, etc.; see also para. 61 where the conversation uses chat, text, video, or voice messaging)
As per dependent Claim 18,
Betteridge/Sullivan teaches the method of claim 17.
Betteridge further teaches:
wherein the communication channels include: short messaging service (SMS), email, web browser, voice call, data call and chat (para. 57 user and agent device communicate with each other using communications that include SMS, email, webchat, VOIP (voice over internet protocol), video, social media, etc.; see also para. 61 where the conversation uses chat, text, video, or voice messaging; see also para. 62, 65 where the user transmits the query to initiate engagement with the contact center agent (“data call”))
Claims 2-6 and 12 are rejected under 35 U.S.C. 103 as being unpatentable over Betteridge et al. (US2024/0104469) in view of Sullivan et al. (US 10,554,817) as applied to claim 1 above, further in view of Arya et al. (US2023/0037894).
As per dependent Claim 2,
Betteridge/Sullivan teaches the method of claim 1.
Betteridge suggests the limitation as the platform provides an evaluation function for users to rate their experience with the platform.
Betteridge/Sullivan does not teach, but Arya teaches:
determining, by the enterprise persona, whether the new workflow is associated with a service that the enterprise wants to provide future end-users of the enterprise (figure 1B and para. 39 the accuracy of the auto-generated conversational dialog flow (“new workflow”) is compared to a threshold, where is the accuracy is equal to or greater than the threshold, the flow is sent to the database to update the flow dataset; para. 38 where the flow dataset includes pre-stored flow dialog networks received at the time of bot creation)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Arya with the motivation of improving the performance of the agent/enterprise persona.
See para. 39 “The system of the present disclosure may thus be able to understand an extent of deviation or performance of the chatbot based on analysis of flow and also be able to update the database with the new conversational flows. The self-learning engine 112 may thus be able to identify the aspects pertaining to intent and flow of dialog between user and the chatbot as well as automatically correct the inconsistencies in the behavior or performance of the chatbot.”
As per dependent Claim 3,
Betteridge/Sullivan/Arya teaches the method of claim 2.
Betteridge teaches:
wherein the enterprise persona is a real person or a virtual person (figure 4 and para. 58 where the user engages with a call center or contact center service and the agent (“enterprise persona”) may be a human or non-human)
As per dependent Claim 4,
Betteridge/Sullivan/Arya teaches the method of claim 2.
Sullivan does teach in Col. 24 Lines 42-57 storing the workflow.
Betteridge/Sullivan does not teach, but Arya teaches:
responsive to determining that the new workflow is associated with a useful service, storing the new workflow as an additional workflow module to be added to the plurality of pre-created workflow modules (figure 1B and para. 39 the accuracy of the auto-generated conversational dialog flow (“new workflow”) is compared to a threshold, where is the accuracy is equal to or greater than the threshold, the flow is sent to the database to update the flow dataset and thus “update the database with the new conversational flows”; para. 38 where the flow dataset includes pre-stored flow dialog networks received at the time of bot creation)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Arya with the motivation of improving the performance of the agent/enterprise persona.
See para. 39 “The system of the present disclosure may thus be able to understand an extent of deviation or performance of the chatbot based on analysis of flow and also be able to update the database with the new conversational flows. The self-learning engine 112 may thus be able to identify the aspects pertaining to intent and flow of dialog between user and the chatbot as well as automatically correct the inconsistencies in the behavior or performance of the chatbot.”
As per dependent Claim 5,
Betteridge/Sullivan/Arya teaches the method of claim 4.
Betteridge does not teach, but Sullivan teaches:
passing on the stored workflow to a scheduler module (Col. 24 Lines 42-57 the system stores the agent and contact workflow; Col. 28 Line 64 to Col. 29 Line 5 to service newly discovered intents, generating and storing agent and contact workflow; Col. 30 Line 33-64 the system generates a user interface to prompt the user whether to deploy the automated service agent so that the agent is invoked to execute the contact workflow)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Sullivan with the motivation of improving the performance of the agent/enterprise persona and reduce the amount of work a human would have to perform.
See Col. 4 Lines 21-44 “reduce the amount of collaborative activity that a particular user must manually perform” and Col. 5 Lines 35-40 “to service a customer contact without human intervention and/or to reduce the amount and/or improve the accuracy of actions a human service agent would otherwise perform manually to execute the contact workflow”.
As per dependent Claim 6,
Betteridge/Sullivan/Arya teaches the method of claim 5.
Betteridge does not teach, but Sullivan teaches:
determining, by the scheduler module, a time to deploy the new workflow (Col. 30 Line 33-64 the system generates a user interface to prompt the user whether to deploy the automated service agent so that the agent is invoked to execute the contact workflow and if the user accepts, the system proceeds with integrating the agent – the user may also choose to modify parameters or choose to not deploy and simply store or delete so the system would perform tasks to abort the deployment)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Sullivan with the motivation of improving the performance of the agent/enterprise persona and reduce the amount of work a human would have to perform. See Col. 4 Lines 21-44 and Col. 5 Lines 35-40.
As per dependent Claim 12,
Betteridge/Sullivan/Arya teaches the method of claim 2.
Betteridge/Sullivan does not teach, but Arya teaches:
responsive to determining that the new workflow is not associated with a useful service, rejecting the new workflow from being added to the plurality of pre-created workflow modules (figure 1B and para. 39 the accuracy of the auto-generated conversational dialog flow (“new workflow”) is compared to a threshold, where if the accuracy is less than the threshold, the flow is sent to the self-learning engine to be “updated” and the self-learning engine may “correct the inconsistencies in the behavior or performance of the chatbot”)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Arya with the motivation of improving the performance of the agent/enterprise persona.
See para. 39 “The system of the present disclosure may thus be able to understand an extent of deviation or performance of the chatbot based on analysis of flow and also be able to update the database with the new conversational flows. The self-learning engine 112 may thus be able to identify the aspects pertaining to intent and flow of dialog between user and the chatbot as well as automatically correct the inconsistencies in the behavior or performance of the chatbot.”
Claims 7-10 are rejected under 35 U.S.C. 103 as being unpatentable over Betteridge et al. (US2024/0104469) in view of Sullivan et al. (US 10,554,817) in view of Arya et al. (US2023/0037894) as applied to claim 4 above, further in view of Raimondo et al. (US2024/0176958).
As per dependent Claim 7,
Betteridge/Sullivan/Arya teaches the method of claim 4.
Betteridge/Sullivan does not teach, but Arya teaches:
further training a chatbot with the new workflow, wherein the chatbot is already pre-trained with the pre-created workflows (para. 37 and 39 the auto-generated responses/dialog flow (“new workflow”) is sent to the database for updating the training data; para. 36 database having prestored training data such as chat logs; see also para. 46-48)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Arya with the motivation of improving the performance of the agent/enterprise persona.
See para. 37 “The system of the present disclosure may thus be able to firstly identify the type of intent and also to generate auto-responses for new intents” and para. 39 “The system of the present disclosure may thus be able to understand an extent of deviation or performance of the chatbot based on analysis of flow and also be able to update the database with the new conversational flows. The self-learning engine 112 may thus be able to identify the aspects pertaining to intent and flow of dialog between user and the chatbot as well as automatically correct the inconsistencies in the behavior or performance of the chatbot.”
Betteridge/Sullivan/Arya does not teach, but Raimondo teaches:
large language model (LLM) (para. 15 large language model can leverage planning for better performing task-oriented dialogues; para. 25-26 workflows are used as training data for the language model)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Raimondo with the motivation of improving the efficiency of the process as the LLMs are able to better adapt to performing new tasks.
See para. 12 “Technological advantages of the trained models disclosed herein include these models being: a) able to generalize to unseen workflows by following the provided plan, and b) able to generalize to unseen actions if they are provided in the plan. In contrast, models which are provided no workflow information or only the workflow name deteriorate on tasks involving new multi-step plans” and para. 18 “Large language models (LLMs) conditioned on plans unseen during training are able to better adapt to performing new tasks in a zero-shot learning setting.”
As per dependent Claim 8,
Betteridge/Sullivan/Arya/Raimondo teaches the method of claim 7.
Betteridge/Sullivan/Arya does not teach, but Raimondo teaches:
wherein the LLM is fine-tuned with industry-specific data (para. 11 fine-tuned to task oriented dialogues; para. 15 and 21 model is fine tuned on tasks in the ABCD dataset; para. 26 natural language processing model is tuned with action plans/workflows as training data; see also para. 18 and 23 LLM; para. 32 ABCD dataset includes instructions for completing workflows (business logic) and ABCD includes workflows, unique actions, slot values, etc.)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Raimondo with the motivation of improving the efficiency of the process as the LLMs are able to better adapt to performing new tasks.
See para. 12 “Technological advantages of the trained models disclosed herein include these models being: a) able to generalize to unseen workflows by following the provided plan, and b) able to generalize to unseen actions if they are provided in the plan. In contrast, models which are provided no workflow information or only the workflow name deteriorate on tasks involving new multi-step plans” and para. 18 “Large language models (LLMs) conditioned on plans unseen during training are able to better adapt to performing new tasks in a zero-shot learning setting.” See also para. 15 where performance is improved.
As per dependent Claim 9,
Betteridge/Sullivan/Arya/Raimondo teaches the method of claim 8.
Betteridge/Sullivan/Arya does not teach, but Raimondo teaches:
wherein the LLM is fine-tuned with task-specific data (para. 11 fine-tuned to task oriented dialogues; para. 15 and 21 model is fine tuned on tasks in the ABCD dataset; para. 26 natural language processing model is tuned with action plans/workflows as training data; see also para. 18 and 23 LLM)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Raimondo with the motivation of improving the efficiency of the process as the LLMs are able to better adapt to performing new tasks.
See para. 12 “Technological advantages of the trained models disclosed herein include these models being: a) able to generalize to unseen workflows by following the provided plan, and b) able to generalize to unseen actions if they are provided in the plan. In contrast, models which are provided no workflow information or only the workflow name deteriorate on tasks involving new multi-step plans” and para. 18 “Large language models (LLMs) conditioned on plans unseen during training are able to better adapt to performing new tasks in a zero-shot learning setting.” See also para. 15 where performance is improved.
As per dependent Claim 10,
Betteridge/Sullivan/Arya/Raimondo teaches the method of claim 9.
Betteridge/Sullivan/Arya does not teach, but Raimondo teaches:
wherein the fine-tuned LLM receives one or more of the following as inputs: workflow variables, contextual information about communication channel, previous conversation history, and guidelines (para. 11 input into the model includes one or more plan steps or a full short example of how the model interacts with the customer; para. 23 where text inputs of the client correspond to inputs to the model; para. 29-30 one or more plan steps can be utilized to guide the model; para. 15 and 21 model is fine-tuned on the tasks in the ABCD dataset and the model augments the dialogue context; para. 36-37 context data items are determined (input context string of the dialogue history) and context is augmented)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Raimondo with the motivation of improving the agent.
See para. 12 “Prior approaches that create dialogue agents based on large language models provided only with conversation history as input do not reliably perform new multi-step tasks that are unseen during training. To address these limitations, the techniques disclosed herein augment dialogue contexts with prompts comprising known workflows. Once an appropriate workflow has been identified, it is possible to inject workflow plans as prompts comprised of action sequences required to accomplish a task, possibly augmented with additional metadata. In such a framework, it is possible to obtain workflow prompts from a symbolic planning mechanism or other types of external API calls. This allows new workflows to be added and intermediate steps to be changed and reassembled with guarantees that a correct and up-to-date plan is provided.”
Claim 11 is rejected under 35 U.S.C. 103 as being unpatentable over Betteridge et al. (US2024/0104469) in view of Sullivan et al. (US 10,554,817) in view of Arya et al. (US2023/0037894) in view of Raimondo et al. (US2024/0176958) as applied to claim 10 above, further in view of Malak et al. (US 20250094703).
As per dependent Claim 11,
Betteridge/Sullivan/Arya/Raimondo teaches the method of claim 10.
Betteridge/Sullivan/Arya/Raimondo does not teach, but Malak teaches:
wherein LLM generates a JSON representation of the new workflow (para. 16 generate a JSON object using a LLM model to create a data visualization responsive to the user’s request; figure 5 and para. 64-70 where in para. 64-65 the user requests a visualization, the application classifies the user input and selects prompt template to generate a prompt (“new workflow”) and in para. 66-67 generating the output in JSON format according to the instructions provided in the prompt template)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with Malak with the motivation of improving the user experience.
See para. 28 “Moreover, by tasking an LLM with making selections to configure the visualization, the process is made more expedient-a single input by the user enables the creation of a visualization which the user can then tweak or modify as necessary to the type of display the user is seeking or to quickly and easily explore alternative ways to visualize the data. By using an LLM to assist in generating the visualization, a wealth of background knowledge in the form of the training data used to train the LLM is made available to the user to figure out the most suitable or most useful visualization for the user's needs.”
Claim 15 is rejected under 35 U.S.C. 103 as being unpatentable over Betteridge et al. (US2024/0104469) in view of Sullivan et al. (US 10,554,817) as applied to claim 1 above, further in view of James et al. (US2022/0309365).
As per dependent Claim 15,
Betteridge/Sullivan teaches the method of claim 1.
Betteridge teaches:
wherein the micro-engagement engine uses a software to analyze the end-user’s input collected from messages (para. 60 analyzes the text for keywords to determine the intent of the query received from the user device and the context of the keywords may be determined using a machine learning model; para. 57 and 61 conversation using chat, text, video, etc.)
Betteridge/Sullivan does not teach, but James teaches:
wherein the micro-engagement engine uses a Document Intelligence Services Architecture (DISA) to analyze the end-user’s input collected from a document provided or a form filled by the end-user (para. 66 where a user may upload a digital document and the AI architecture may perform analysis of the document in para. 67-71; see also para. 29-31 where the user may transmit a request including digital documents)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify the Betteridge invention with James since each individual element and its function are shown in the prior art, albeit shown in separate references, the difference between the claimed subject matter and the prior art rests not on any individual element or function but in the very combination itself. That is in the substitution of DISA of James for the software of Betteridge. Both are elements or techniques utilized for analysis of the user’s input. Further Betteridge teaches a plurality of ways in which user input is collected and James teaches input collected from a document or form. Thus, the simple substitution of one known element for another producing a predictable result renders the claim obvious.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Tsuchiyma et al. (US2023/0385939)
Liao et al. (US2021/0295203)
Hecht et al. (US2024/0403328)
Wang et al. (US2021/0350209)
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 Lisa Ma whose telephone number is (571)272-2495. The examiner can normally be reached Monday to Thursday 7 AM - 5 PM.
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, Shannon Campbell can be reached at (571)272-5587. 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.
/L.M./Examiner, Art Unit 3628
/SHANNON S CAMPBELL/Supervisory Patent Examiner, Art Unit 3628