DETAILED ACTION
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)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(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.
Claims 1, 17, and 20 are rejected under 35 U.S.C. 102(a)(1) and 102(a)(2) as being anticipated by U.S. Patent Application Publication No. 2021/0201144 A1 to Siddhartha R. Jonnalagadda et al. (hereinafter Gadda).
Regarding claim 1, Gadda teaches a method for data processing at a multi-tenant generative artificial intelligence (AI) system, comprising: receiving a configuration associated with a tenant of the multi-tenant generative AI system and tenant-specific training data associated with the tenant, (Gadda teaches receiving data associated with a user to provide the AI system with a target database for message generation (i.e., training data). Gadda at ¶¶ [0108] - [0111]. Further, Gadda teaches a user inputting a configuration using preferences and settings for the conversation system.)
wherein the configuration includes a first indication of a first communication channel over which a tenant-specific conversational agent is to communicate with users associated with the tenant and wherein the tenant-specific training data includes context information associated with the tenant that is expressed in natural language; (Gadda teaches the user inputting a preferred method of communication (i.e., channel) which includes a variety of possible communication channels. Gadda at ¶¶ [0127] - [0129].)
determining an intent of a query received from the tenant based at least in part on an analysis of the query; (Gadda teaches determining the intent of a message using artificial intelligence models. Gadda at ¶¶ [0097] - [0098].)
transmitting the query to a first generative AI model of a plurality of generative AI models, wherein the first generative AI model is selected based at least in part on the determined intent; (Gadda teaches selecting a model and classifying the input based on the confidence levels which are based on the intent. Gadda at ¶¶ [0097] - [0098].)
and transmitting, to the tenant over the first communication channel, a response to the query generated by the first generative AI model. (Gadda teaches building and sending an outgoing message using a message builder. Gadda at ¶¶ [0087] - [0091].)
Regarding claim 20, Gadda teaches a multi-tenant generative artificial intelligence (AI) system for data processing, comprising: one or more memories storing processor-executable code; and one or more processors coupled with the one or more memories and individually or collectively operable to execute the code to cause the multi-tenant generative artificial intelligence (AI) system to: (Gadda teaches the system implemented on a computer comprising a processor, and memories, as well as other common components of a computer system. Gadda at ¶¶ [0193] - [0195].)
receive a configuration associated with a tenant of the multi-tenant generative AI system and tenant-specific training data associated with the tenant, (Gadda teaches receiving data associated with a user to provide the AI system with a target database for message generation (i.e., training data). Gadda at ¶¶ [0108] - [0111]. Further, Gadda teaches a user inputting a configuration using preferences and settings for the conversation system.)
wherein the configuration includes a first indication of a first communication channel over which a tenant-specific conversational agent is to communicate with users associated with the tenant and wherein the tenant-specific training data includes context information associated with the tenant that is expressed in natural language; (Gadda teaches the user inputting a preferred method of communication (i.e., channel) which includes a variety of possible communication channels. Gadda at ¶¶ [0127] - [0129].)
determine an intent of a query received from the tenant based at least in part on an analysis of the query; (Gadda teaches determining the intent of a message using artificial intelligence models. Gadda at ¶¶ [0097] - [0098].)
transmit the query to a first generative AI model of a plurality of generative AI models, wherein the first generative AI model is selected based at least in part on the determined intent; (Gadda teaches selecting a model and classifying the input based on the confidence levels which are based on the intent. Gadda at ¶¶ [0097] - [0098].)
and transmit, to the tenant over the first communication channel, a response to the query generated by the first generative AI model. (Gadda teaches building and sending an outgoing message using a message builder. Gadda at ¶¶ [0087] - [0091].)
Regarding claim 17, Gadda teaches all the limitations of claim 1 as laid out above. Further, Gadda teaches the method of claim 1, further comprising: modifying one or more elements of the configuration, the tenant-specific training data, or both, based at least in part on reception of one or more modification instructions received via a tenant portal. (Gadda teaches the user configuring their preferences and settings (i.e., elements of the configuration) and alerting the user for manual intervention. (i.e., the configuration is performed manually, through some sort of interaction, or "portal" persay.) Gadda at ¶¶ [0111].)
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.
Claims 2 – 5, 7 – 11, 13, and 18 – 19 are rejected under 35 U.S.C. 103 as being unpatentable over Gadda in view of U.S. Patent Application Publication No. 2025/0284893 A1 to Martin Neale et al. (hereinafter Neale).
Regarding claim 2, Gadda teaches all the limitations of claim 1. Gadda, however, does not teach the limitations of claim 2.
In a similar field of endeavor (e.g., client/user specific models trained based upon information received by the user wherein the models are customized/personalized based on configurations received by the system.), Neale teaches the method of claim 1, further comprising: receiving tenant-specific services data associated with the tenant; (Neale teaches receiving data associated with client-specific instances of the virtual assistant platform (i.e., generative AI system). Neale at ¶¶ [0081] - [0088].)
and modifying one or more parameters of one or more elements of the multi-tenant generative AI system based at least in part on the tenant-specific services data. (Neale teaches updating the trained AI model based upon the received client-specific information. Neale at ¶¶ [0081] - [0088].)
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date to combine the teachings of Gadda with the teachings of Neale to provide the limitations of claim 3. Doing so would have improved training and improved the virtual assistant (i.e., generative AI) by providing multiple assistants in multiple contexts which are further improved by increasing success of the AI assistants in analogous conversations with humans as recognized by Neale at ¶¶ [0059] - [0060], and [0068].
Regarding claim 3, Gadda teaches all the limitations of claim 1 as laid out above. Gadda, however, does not teach all the limitations of claim 3. In a similar field of endeavor (e.g., client/user specific models trained on user specific data), Neale teaches the method of claim 1, further comprising: parsing the tenant-specific training data to determine the context information, one or more actions that are available to the tenant, one or more conditions associated with the one or more actions, one or more procedures associated with the tenant, one or more services associated with the tenant, or any combination thereof. (Neale teaches the client-specific data including information related to the context of a response, the possible actions associated with the client (e.g., checking your balance or pay your rent, etc.) and services associated with the client (i.e., rent management/housing account management.) Neale at ¶¶ [0081] - [0088].)
Regarding claim 4, Gadda teaches all the limitations of claim 1, as laid out above. Gadda, however, does not teach all the limitations of claim 4.
In a similar field of endeavor, (e.g., client/user specific models trained on user specific data), Neale teaches the method of claim 1, wherein determining the intent of the query is based at least in part on the tenant-specific training data. (Neale teaches identifying the intent of a query from a user by processing the query using the trained AI models. Neale at ¶¶ [0081] - [0083].)
Regarding claim 5, Gadda teaches all the limitations of claim 1, as laid out above. Gadda, however, does not teach all the limitations of claim 5.
In a similar field of endeavor (e.g., client/user specific models trained on user specific data), Neale teaches the method of claim 1, further comprising: transmitting, to a service associated with the multi-tenant generative AI system, a request to perform an action indicated in the response to the query; (Neale teaches invoking actions in response to a user query, as well as including information associated with the action in a response to the query issued to the user. Neale at ¶¶ [0081] - [0087].)
receiving, from the service, a first indication that the action has been performed; (Neale teaches sending to the user, an indication that an action has been invoked or completed wherein the query was sent to a system separate from the user's device and the indication was transmitted back to the user's device. Neale at ¶¶ [0081] - [0087] and Fig. 4.)
and transmitting, to the tenant over the first communication channel, a second indication that the action has been performed. (Neale teaches sending to the user, an indication that an action has been invoked or completed. Neale at ¶¶ [0081] - [0087].)
Regarding claim 7, Gadda in view of Neale (hereinafter Gadda-Neale) teaches all the limitations of claim 5 as laid out above. Further, Neal teaches the method of claim 5, further comprising: selecting the service based at least in part on the configuration, the tenant-specific training data, or both. (Neale teaches performing specific actions in response to user queries wherein the actions are specific to the organization for which the model was trained (i.e., the action selected in response to the user query and intent is based on the configuration and client specific training of the model). Neale at ¶¶ [0081] - [0087].)
Regarding claim 8, Gadda-Neale teaches all the limitations of claim 5 as laid out above. Further, Neale teaches the method of claim 5, further comprising: detecting a state of the service based at least in part on the configuration, the tenant-specific training data, or both, wherein the transmission of the request to perform the action is based at least in part on the detection of the state. (Neale teaches detecting an anomaly of the AI virtual assistant (i.e., a state of the assistant) and preventing further action to update the model and fix the anomaly (i.e., the transmission of the request to perform the action is based on the detection of the state.) Neale at ¶¶ [0112] - [0113].)
Regarding claim 9, Gadda teaches all the limitations of claim 1, as laid out above. Gadda, however, does not teach all the limitations of claim 9. In a similar field of endeavor (e.g., client/user specific models trained on user specific data), Neale teaches the method of claim 1, further comprising: validating the response to the query via a retrieval augmented generation (RAG) service, or an in-context learning (ICL) service, or both, the validation being based at least in part on the configuration, the tenant-specific training data, or both. (Neale teaches updating the context and intent database based on the response due to anomalies within the AI system. (i.e., in context learning is performed and the response is not validated because the context and intent must be altered.) Neale at ¶¶ [0099] - [0104].)
Regarding claim 10, Gadda teaches all the limitations of claim 1 as laid out above. Gadda, however, does not teach all the limitations of claim 10. In a similar field of endeavor (e.g., client/user specific models trained on user specific data), Neale teaches the method of claim 1, further comprising: modifying one or more parameters associated with the first generative AI model based at least in part on the determined intent. (Neale teaches a mesh learning process that harvests anonymized data from all AI assistants and learns based off of the data. Neale at ¶¶ [0081] - [0087]. Further, Neale teaches AI is used to analyze data to determine anomalies that include intents that need to be replaced (i.e., determined intents require updating of the models.) Neale at ¶¶ [0081] - [0087].)
Regarding claim 11, Gadda teaches all the limitations of claim 1 as laid out above. Gadda, however, does not teach all the limitations of claim 11. In a similar field of endeavor (e.g., client/user specific models trained on user specific data), Neale teaches the method of claim 1, further comprising: training the first generative AI model based at least in part on the tenant-specific training data, an analysis of one or more communications with the tenant-specific conversational agent, one or more previous responses generated by the first generative AI model, or any combination thereof. (Neale teaches training AI models based on client-specific data (i.e., tenant specific data) which includes training data that comprises annotated queries and responses including queries associated with intent labels. (i.e., one or more communications, and previous responses). Neale at ¶¶ [0007] - [0016] and [0061] - [0062].)
Regarding claim 13, Gadda teaches all the limitations of claim 1 as laid out above. Gadda, however, does not teach all the limitations of claim 13. In a similar field of endeavor (e.g., client/user specific models trained on user specific data), Neale teaches the method of claim 1, further comprising: presenting, via a user interface, a graphical representation of information indicated in the response to the query, wherein the graphical representation is presented in a tenant-specific format determined based at least in part on the query, the configuration, the tenant-specific training data, or any combination thereof, and wherein the tenant-specific format is specified in the configuration. (Neale teaches presenting information in a user interface that is displayed in the form of a chat-room/session (i.e., a graphical representation of the information in the response.) Neale at ¶ [0061]. Further, the output responses are determined by the client-specific trained models. Neale at ¶¶ [0081] - [0087]. Which means, in essence, that the format and configuration of the responses is client-specific.)
Regarding claim 18, Gadda teaches all the limitations of claim 1 as laid out above. Gadda, however, does not teach all the limitations of claim 18. In a similar field of endeavor (e.g., client/user specific models trained on user specific data), Neale teaches the method of claim 1, further comprising: transmitting, to the tenant over the first communication channel, tenant-specific reporting associated with the tenant and the multi-tenant generative AI system. (Neale teaches constant monitoring and updating of language models and receiving client approval of updates, when the updates outcome are positive, before implementing the updates (i.e., the client receives some form of information/reporting regarding the outcome of the update in order to provide approval.) Neale at ¶¶ [0114] - [0115].)
Regarding claim 19, Gadda teaches all the limitations of claim 1 as laid out above. Gadda, however, does not teach all the limitations of claim 19. In a similar field of endeavor (e.g., client/user specific models trained on user specific data), Neale teaches the method of claim 1, further comprising: determining a tenant-specific processing flow that is associated with the tenant and that comprises operations associated with the first generative AI model, one or more additional generative AI models of the plurality of generative AI models, or any combination thereof; (Neale teaches sets of client specific models operating within specific contexts that are trained for specific contexts based on specific locations (i.e., a specific processing flow/context associated with specific locations of operating AI models for the clients.) Neale at ¶¶ [0021] - [0024].)
wherein transmitting the query to the first generative AI model is based at least in part on the tenant-specific processing flow. (Neale teaches sending queries to a model to invoke actions and providing a response to the query including information about the invoked action. Neale at ¶¶ [0081] - [0087].)
Claim 6 is rejected under 35 U.S.C. 103 as being unpatentable over Gadda-Neale as applied to claim 5 above, and further in view of U.S. Patent Application Publication No. 2023/0245651 A1 to Jenny Z. Wang (hereinafter Wang).
Regarding claim 6, Gadda-Neale teaches all the limitations of claim 5 as laid out above. Gadda-Neale, however, do not teach all the limitations of claim 6.
In a similar field of endeavor (e.g., adapting a conversational AI to specific users and authentication of users for AI use), Wang teaches the method of claim 5, further comprising: requesting one or more credentials from a credential repository based at least in part on the response to the query, wherein the request to perform the action comprises an indication of the one or more credentials. (Wang teaches verifying and authenticating a user for access to specific APIs and services associated with an AI model. (i.e., requesting the credentials used for user authentication of any type as discussed by Gadda at ¶ [0109]). Wang at ¶¶ [0183] - [0185] and [0218] - [0227].)
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date to combine the teachings of Gadda-Neale with the teachings of Wang to provide the limitations of claim 6. Doing so would have improved accuracy and efficiency of the models as well as improved user experience by considering individual preferences as recognized by Wang at ¶¶ [0033] - [0034]. Further, Wang's additional would improve response times and customer satisfaction as recognized by Wang at ¶ [0037].
Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over Gadda as applied to claim 1 above in view of U.S. 2013/0086027 A1 to Matthieu Hebert (hereinafter Hebert).
Regarding claim 12, Gadda teaches all the limitations of claim 1 as laid out above. Gadda, however, does not teach all the limitations of claim 12.
In a similar field of endeavor (e.g., AI models for processing user queries and answering questions), Hebert teaches the method of claim 1, further comprising: transmitting, to the tenant over the first communication channel, an indication of a suggested modification to the query based at least in part on the analysis of the query, the tenant-specific training data, or both; and modifying the query based at least in part on reception of a response to the indication of the suggested modification to the query. (Hebert teaches determining alternate suggestions to provide to the user to alter their query and modifying the query and resubmitting the query based on the suggestions. Hebert at ¶ [0129].)
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date to combine the teachings of Gadda with the teachings of Wang to provide the limitations of claim 12. Doing so would have allowed the user to improve their queries without increasing cognitive load as recognized by Hebert at ¶¶ [0128] - [0129].
Claims 14 - 15 are rejected under 35 U.S.C. 103 as being unpatentable over Gadda in view of U.S. Patent Application Publication No. 2023/0252224 A1 to Bao Tran (hereinafter Tran).
Regarding claim 14, Gadda teaches all the limitations of claim 1 as laid out above. Gadda, however, does not teach all the limitations of claim 14.
In a similar field of endeavor (e.g., generative AI models generating text responses), Tran teaches the method of claim 1, wherein the configuration further includes a second indication of a second communication channel associated with escalation operations and a third indication of moratorium information associated with operation of the tenant-specific conversational agent. (Tran teaches receiving a request to escalate the interaction between an agent and a user to a supervisor, as well as placing users on hold (i.e., transmitting an indication of a moratorium). Tran at ¶¶ [0238] - [0255].)
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date to combine the teachings of Gadda with the teachings of Tran to provide the limitations of claim 14. Doing so would have would have allowed the system to order users by those who have been placed on a moratorium the longest which would allow the system to provide optimal interactions between users and agents during escalations as recognized by Tran at ¶¶ [0238] - [0255].
Regarding claim 15, Gadda in view of Tran (hereinafter Gadda-Tran) teaches all the limitations of claim 14. Further, Tran teaches the method of claim 14, further comprising: transmitting an escalation request via the second communication channel based at least in part on an escalation indication comprised in the response to the query. (Tran teaches the system selecting and transferring the user to a different agent or supervisory (i.e., an indication of escalation is given by transferring the user to a different agent or supervisor.) Tran at ¶¶ [0238] - [0255].)
Claim 16 is rejected under 35 U.S.C. 103 as being unpatentable over Gadda-Tran as applied to claims 14 and 15 above, and further in view of U.S. Patent Application Publication No. 2022/0180874 A1 to Livio Pugliese et al. (hereinafter Pugliese).
Regarding claim 16, Gadda-Tran teaches all the limitations of claim 14 as laid out above. Gadda-Tran, however, do not teach all the limitations of claim 16.
In a similar field of endeavor (e.g., chatbots and speech processing of queries), Pugliese teaches the method of claim 14, further comprising: transmitting, to the tenant over the first communication channel and based at least in part on satisfaction of a moratorium condition indicated in the moratorium information, an indication of a moratorium period, wherein the indication of the moratorium period comprises a length of the moratorium period, one or more indications of prioritized operations associated with the moratorium period, or any combination thereof. (Pugliese teaches transferring the user to speak with an agent wherein the user is informed by a chatbot "please hold while I get the operator." Pugliese at ¶ [0140]. This includes both an indication of a moratorium (i.e., Please hold) and an indication of the length of the moratorium (i.e., while I get the operator). Pugliese at ¶ [0140].)
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date to combine the teachings of Gadda-Tran with the teachings of Pugliese to provide the limitations of claim 16. Doing so would have would have provided improved chatbot functionality and implement a more natural and flexible conversation between chatbots and users as recognized by Pugliese at ¶¶ [0010] - [0013].
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CAMERON KENNETH YOUNG whose telephone number is (703)756-1527. The examiner can normally be reached Mon - Fri, 9:00 AM - 5:00 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, Andrew Flanders can be reached at 571-272-7516. 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.
/CAMERON KENNETH YOUNG/ Examiner, Art Unit 2655
/ANDREW C FLANDERS/ Supervisory Patent Examiner, Art Unit 2655