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 .
DETAILED ACTION
This Office Action is in response to the application filed on 12/19/2024. Claims 1 and 11 are independent claims. Claims 1-20 have been examined and are pending. This Action is made non-FINAL.
Drawings
The drawings were received on 12/19/2024. These drawings are reviewed and accepted by the Examiner.
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 1-4, 6-10, 11-14, and 16-20 are rejected under 35 U.S.C. 103 as being unpatentable over O’Hern et al. (“O’Hern,” US 026/0080067) in view of Reagan et al (“Reagan,” US 2025/0363200),
Regarding claim 1, O’Hera teaches a method for managing requests, the method comprising:
intercepting, by an inspection module, a request, wherein the request was initiated by a user using a user interface (O’Hern: par. [0067], "the method 500 may comprise processing (e.g. by the processing device) the AI servicer request through AI enclave may instead intercept, filter, direct, redirect, quarantine, and/or otherwise alter the communication path between the user and the requested AI service provider/device". O'Hern further teaches receiving requests from a user device/user interface for execution of an AI service (pars. [0052], [0066]);
extracting, by the inspection module, content from the request, wherein the content is sent to a classification module (O’Hern: par. [0034], "content moderation program 242c may , for example, execute dual LLM instances to derive, identify, and/or utilize one or more LLM instances to derive, identify, and/or categorize an intent or content of an outgoing request with respect to stored rules, thresholds, criteria, etc.".);
classifying, by the classification module, the content (O’Hern: par. [0034], "derive, identify, and/or categorize an intent or content of an outgoing request");
determining, by the classification module and based on the classifying, that the request violates an organization’s policy (O’Hern: par. [0035], "where the content of a request is determined to violate a rule or threshold". O'Hern further teaches evaluation against: "stored rules, thresholds, criteria" (par. [0034]));
in response to the determining, modifying, by the classification module, the request to generate a modified request (O’Hern: par. [0041], “and requests violating the model may be either rejected (e.g., prevented from being sent to the external system 306) and/or modified (e.g., dynamically) to remove and/or replace the offending content and/or intent.. O'Hern further teaches: "automatically replace sensitive data with test data and/or fake data" (par. [0041]));
by the classification module, the modified request (O’Hern: par. [0041], request violating the model may be … modified to remove and/or replacing offending content and/or intent) to an artificial intelligence (AI) service (O’Hern: par. [0074], In a case where the AI service request comprises a request for sample code that accomplishes a particular task, for example, the AI enclave may automatically identify an AI service that specializes in (and/or has a historic success rate with) coding requests...”; pars. [0052]-[0067]), wherein the AI service corresponds to the user interface. (O’Hern: pars. [0052]-[0067]);
receiving, in response to the modified request and by the inspection module, a response from the AI service, wherein the response is provided to the user interface (O'Hern teaches receiving AI-generated responses and returning such responses through the AI enclave to the user device and interface (pars. [0059]-[0061])); and
displaying, by the user interface, the response to the user (O’Hern: teaches providing the AI-generated response to the user device and interface for presentation to the user (pars. [0060]-[0061])).
O'Hern teaches by the classification module, the modified request to an artificial intelligence (AI) service, wherein the AI service corresponds to the user interface but does not explicitly
issuing the modified request to an artificial intelligence (AI) service
However, in an analogous art, Reagan teaches issuing, the modified request to an artificial intelligence (AI) service (Reagan: par. [0015], route the modified second input to an alternate model. Figure 7: Modified Input With Prompt [Wingdings font/0xE8] Destination Mode; par. 0169, at stage 760 a modified input with the injected prompt [], the destination model 775 can be an LLM and the modified input can include a query; Figure 6: Model Routing [Wingdings font/0xE8] AI Service; par. [0153] At stage 640, the management rules and remediation scores can dictate whether to route the modified query to an alternative model, such as model 2 at AI service);
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Reagan with the method and system of O’hern to include issuing, the modified request to an artificial intelligence (AI) service. One would have been motivated to do so because Reagan expressly teaches routing modified queries and modified inputs to destination AI models after rule evaluation, remediation scoring, prompt injection, and transformation (Reagan: pars. [0153], [0169]). Such incorporation would have predictably enabled O'Hern's moderated requests to continue being processed by AI services while maintaining compliance with organizational rules and content restrictions.
Regarding claim 2, the combination of O’Hern and Reagan teaches the method of claim 1, The combination of O’Hern and Reagan further teaches the modified request is generated in a manner that is transparent to the user (O’Hern: par. [0041], and requests violating the model may be either rejected (e.g., prevented from being sent to the external system 306) and/or modified (e.g., dynamically) to remove and/or replace the offending content and/or intent []automatically replace the sensitive data with test data and/or fake data; par. [0061], “In some embodiments, the method 400 may be transparent to the user The user may submit the AI service request/input (e.g., at "3") and receive the AI response/output (e.g., at "31"), for example, without knowledge and/or involvement in the processes (e.g., AI model execution(s)) of the AI enclave 410”; Reagan: pars. [0153], [0169]).
Regarding claim 3, the combination of O’Hern and Reagan teaches the method of claim 2. The combination of O’Hern and Reagan further teaches
wherein the modified request triggers the AI service to generate the response (O'Hern: pars. [0059]-[0061], requests are transmitted to AI service 406 and AI responses are returned to the user; par. [0041], requests violating the model may be modified to remove and/or replace offending content and/or intent ; Reagan teaches "route the modified query to an alternative model" (par. [0153]). And "modified input ... supplied to the destination model" (par. [0169])), and
wherein the response specifies that the request violates the organization’s policy (O'Hern: par. [0035], "where content of a request is determined to violate a rule or threshold, the AI enclave may provide a response explaining why the request cannot be forwarded").
Regarding claim 4, the combination of O’Hern and Reagan teaches the method of claim 1. The combination of O’Hern and Reagan further teaches comprising:
intercepting a second request, wherein the second request was initiated by the user using a second user interface (O'Hern: pars. [0067], [0052], [0066]);
extracting second content from the second request (O’Hern: par. [0034], "derive, identify, and/or categorize an intent or content of an outgoing request.");
classifying the second content (O’Hern: par. [0034], "categorize an intent or content.");
determining, based on the classifying of the second content, that the second request does not violate the organization’s policy (O’Hern: par. [0035], "where the content of a request is determined to violate a rule or threshold"). O'Hern further teaches evaluation against: "stored rules, thresholds, criteria" (par. [0034]).
in response to the determining that the second request does not violate the organization’s policy, issuing the second request to a second AI service, wherein the second AI service corresponds to the second user interface (O'Hern: teaches communication between a user and a requested AI service provider/device through the AI enclave (pars. [0061], [0067]), and teaches that requests violating rules or thresholds are not forwarded (par. [0035]. Reagan further teaches that management rules and remediation scores dictate whether to "route the modified query to an alternative model" (par. [0153])).
receiving, in response to the second request, a second response from the second AI service (O’Hern: pars. [0059]-[0061], receiving AI-generated responses and returning such responses through the AI enclave to the user device and interface); and
displaying, by the second user interface, the second response to the user (O’Hern: par. [0061], response provided to user device/interface as output.).
Regarding claim 6, the combination of O’Hern and Reagan teaches the method of claim 1. The combination of O’Hern and Reagan further teaches, wherein the content is source code (O’Hern: par. [0074], In a case where the AI service request comprises a request for sample code that accomplishes a particular task, for example, the AI enclave may automatically identify an AI service that specializes in (and/or has a historic success rate with) coding requests...”; par. [0045], “the code improvement and management 342e may comprise analysis tools and/or routines that examine source code (e.g., static code) and/or executed code (e.g., dynamic code)”).
Regarding claim 7, the combination of O’Hern and Reagan teaches the method of claim 1, The combination of O’Hern and Reagan further teaches wherein the AI service is a generative AI service (O’Hern: par. [0074], In a case where the AI service request comprises a request for sample code that accomplishes a particular task…”; par. [0045], "the response from the AI service 306a ... comprises one or more portions of code".).
Regarding claim 8, the combination of O’Hern and Reagan teaches the method of claim 1. The combination of O’Hern and Reagan further teaches the method of claim 1
wherein the modified request triggers the AI service to generate the response (O'Hern: par. [0073]: AI service request may be modified; par. [0074], the AI service request is transmitted to the requested AI service; par. [0075], AI service generates and returns a response (we discussed this flow previously; Reagan: par. [0153], route the modified query to an alternative model; par. [0169], modified input supplied to destination model), and
wherein the response specifies that the request violates the organization’s policy (O’Hern: par. [0035], where content violates a rule or threshold, the AI enclave may provide a response explaining why the request cannot be forwarded) and provides an alternate source from which the user is able to obtain information related to the request (O'Hern: par. [0074], the AI service request may alternatively be routed and/or rerouted to a different AI service provider and the AI enclave may automatically identify an AI service [] that specializes in coding requests []and prioritize internal services over external services; Reagan: par. [0153], route modified query to an alternative model).
Regarding claim 9, the combination of O’Hern and Reagan teaches the method of claim 1. The combination of O’Hern and Reagan further teaches
wherein the inspection module, the classification module, and the user interface are deployed to a computing device (O'Hern: par. [0030], content moderation program 242c and risk-based access control program 242d stored and executed by processing devices; par. [0119], modules, scripts, models, and routines executed by processing devices; pars. [0061], [0073]-[0074], user interface and user device interactions),
wherein the computing device executes in an information technology environment that is related to the user (O'Hern: pars. [0045], [0046], [0074], enterprise/company coding guidelines, enterprise/company data, internal systems, trusted company AI services.).
Regarding claim 10, the combination of O’Hern and Reagan teaches the method of claim 1. The combination of O’Hern and Reagan further teaches
wherein the inspection module and the user interface are deployed to a first computing device of an information technology environment that is related to the user (O’Hern: par. [0067], “the AI enclave may intercept, filter, direct, redirect, quarantine, and/or otherwise alter the communication path between the user and the requested AI service provider/device”; par. [0061], “the user may submit the AI service request/input ... and receive the AI response/output" through the user device/interface..”; Fig. 1; pars. [0019], [0027], user devices 102a-n communicate through network 104..),
wherein the classification module is deployed to a second computing device of the information technology environment (O'Hern: par. [0030], content moderation program 242c and risk-based access control program 242d; par. [0034], deriving, identifying, and categorizing content and intent of requests; Fig. 1; pars. [0019], [0027], [0028]; server device 106, AI device 110, user devices 102a-n, and network 104).
Regarding claim 11, the combination of O’Hern and Reagan teaches a system for managing requests, the system comprising:
a classification module (O’Hern: pars. [0030], [0034]);
a user interface (O’Hern: pars. [0061], [0074]); and
an inspection module (O’Hern: par. [0067]), wherein the classification module, the user interface, and the inspection module are operatively connected to each other over a network (O’Hern: fig. 1. par. 0019),
wherein the inspection module comprises a processor comprising circuitry and memory comprising instructions (O’Hern: fig. 1, pars. [0019], [0027], [0028], [0119]), which when executed by the processor perform associated with the method claimed in claim 1; claim 11 is similar in scope to claim 1, and is therefore rejected under similar rationale.
Regarding claim 12, claim 12 is similar in scope to claim 2, and is therefore rejected under similar rationale.
Regarding claim 13, claim 13 is similar in scope to claim 3, and is therefore rejected under similar rationale.
Regarding claim 14, claim 14 is similar in scope to claim 4, and is therefore rejected under similar rationale.
Regarding claim 16, claim 16 is similar in scope to claim 6, and is therefore rejected under similar rationale.
Regarding claim 17, claim 17 is similar in scope to claim 7, and is therefore rejected under similar rationale.
Regarding claim 18, claim 18 is similar in scope to claim 8, and is therefore rejected under similar rationale.
Regarding claim 19, claim 19 is similar in scope to claim 9, and is therefore rejected under similar rationale.
Regarding claim 20, claim 20 is similar in scope to claim 20, and is therefore rejected under similar rationale.
Claims 5 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over O’Hern et al. (“O’Hern,” US 026/0080067) in view of Reagan et al (“Reagan,” US 2025/0363200) , further in view of Mityagin (“Mityagin,” US 2016/0092683).
Regarding claim 5, the combination of O’Hern and Reagan teaches the method of claim 4. O'Hern teaches content moderation and policy evaluation of request content, including determining whether request content violates organizational rules, thresholds, and criteria (O’Hern: pars. [0034]-[0035]). O'Hern further teaches both violation and non-violation processing paths, including modifying requests that violate policies and allowing compliant requests to proceed to AI services (O’Hern: pars. [0035], [0041], [0059]-[0067]). Reagan teaches that "multiple language model requests can be handled simultaneously across different nodes or containers" and that "caching can also be used to provide outputs that recently have been received from the destination model for the same or semantically similar inputs" (Reagan: par. [0079]).
Neither O'Hern nor Reagan explicitly teaches the content and the second content are the same while being subject to different policy outcomes.
However, in an analogous art, Mityagain teaches that "the administrator can specify many different policies for the same content item type and/or many different policies for different content item types" (Mityagain: par. [0053]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Mityagain with the method and system of O’Hern and Reagan to include “the content and the second content are the same. One would have been motivated to apply the policy management techniques of Mityagin to the AI request policy evaluation systems of O'Hern and Reagan in order to permit different policy outcomes for the same content under different organizational policies, rules, or contexts. Mityagin expressly teaches supporting multiple policies for the same content item type (Mityagin: par. [0053]). It would have been obvious to incorporate Mityagin's policy management techniques into the policy evaluation systems of O'Hern and Reagan in order to support multiple policies for the same content type and thereby improve policy administration flexibility.
Regarding claim 15, claim 15 is similar in scope to claim 5, and is therefore rejected under similar rationale.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CANH LE whose telephone number is (571)270-1380. The examiner can normally be reached on Monday to Friday 6:00AM to 3:30PM other Friday off.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Luu Pham, can be reached at telephone number 571-270-5002. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from Patent Center and the Private Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from Patent Center or Private PAIR. Status information for unpublished applications is available through Patent Center and Private PAIR for authorized users only. Should you have questions about access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
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) Form at https://www.uspto.gov/patents/uspto-automated- interview-request-air-form.
/Canh Le/
Examiner, Art Unit 2439
June 22nd, 2026
/LUU T PHAM/Supervisory Patent Examiner, Art Unit 2439
Prosecution Insights