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 Objections
Claim(s) 11 is/are objected to because of the following informalities:
“validating the request in a first trusted computing environment” should read “validating the request in the first trusted computing environment”. Appropriate correction is required.
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.
Claim(s) 11-19 is/are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. The claim(s) does/do not fall within at least one of the four categories of patent eligible subject matter.
Regarding claim 11, the claim does not clearly recite the hardware entity within the claimed system that validates the request and applies the algorithm to the protected information. Although the trusted computing environments may perform these functions, the claim merely recites a first or second “trusted computing environment” without positively reciting hardware component, such as a memory or a hardware processor. Furthermore, the trusted computing environments are not necessarily limited to hardware and may be implemented entirely in software, for example, as a virtual memory or a software sandbox. Accordingly, the claim encompasses software per se and is non-statutory under 35 USC § 101.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-2, 7, 10-12, 16 and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over OHTA et al., US-20240179109-A1 (hereinafter “OHTA ‘109”) in view of Ponsini et al., US-20230351028-A1 (hereinafter “Ponsini ‘028”) and IMMS et al., US-20230305865-A1 (hereinafter “IMMS ‘865”) and Chalk et al., US-20230107510-A1 (hereinafter “Chalk ‘510”).
Per claim 1 (independent):
OHTA ‘109 discloses: A computerized method for quality and permission controlled request qualification comprising:
receiving a user request, wherein the user request is a natural language request, and wherein the request includes a credential token;
validating the request, wherein the validating utilizes the credential token to determine if the user request is rejected or approved
(FIG. 6, [0180], the command generation process (based on a user request) including the guiding control process and output control performed in the chatbot server 30; [0182], First, when detecting that a user terminal device 110 has accessed the chatbot service through the communication control unit 310 (step S101) (as illustrated in FIGS. 4-5, the user and the chatbot exchange natural language thorough an interactive dialogue), the chatbot management unit 211, based on the login scenario data, performs a login process that executes the user authentication to log in the chatbot service (Step S102) (receiving a user request, wherein the user request is a natural language request); [0183], the chatbot management unit 211 identifies the user who accesses the chatbot service as a specific user based on the information such as a login ID (a credential token) when executing the login (Step S104) (validating the request, wherein the validating utilizes the credential token to determine if the user request is rejected or approved); [0184], the selection processing unit 215 ... selects the scenario data and the server system 50 based on the user ID of the specific user and the input text that is entered and recognized (Step S105); [0186], The command generation processing unit 216 performs a command output control processing (for example, determining whether the user request is an email service or a conference room reservation service as shown in FIGS. 4-5) that outputs the generated command to the selected server system 50 ... (Step S107); FIG. 1 and 3, [0099]-[0102], the chatbot server 30 performs operations including “recognize the input text entered by a user,” “identify the user who enters the text as a specific user,” and “select the scenario data from among a plurality of scenario data and to select the server system from a plurality of server systems.” For selecting the service scenario to be executed, the chatbot server uses both the user’s login ID and the text entered by the user (See FIGS. 3-5). Accordingly, the user’s request may be interpreted as including not only the context text entered by the user but also the user’s login credentials (e.g., login ID and password));
when the user request is approved, qualifying the user request to a qualified request by matching the user request to one of a plurality of templates; and
applying an algorithm to protected information responsive to the qualified request
(FIG. 6, [0183], the chatbot management unit 211 identifies the user who accesses the chatbot service as a specific user based on the information such as a login ID when executing the login (Step S104; according to FIG. 3, reaching this step indicates that the user request is approved); [0119], In the scenario DB 332, two or more scenario data (a plurality of templates) is stored in association with IDs to identify each scenario data ... to allow the user to enjoy the each of the services; [0122], The "scenario" has a plurality of scripts to provide the service desired by a user, and defines the content and order of each scrips, as well as a type of data required to execute the scripts and content of the data – including an algorithm; [0184], the selection processing unit 215 ... selects the scenario data and the server system 50 based on the user ID of the specific user and the input text that is entered and recognized (Step S105) (qualifying the user request to a qualified request by matching the user request to one of a plurality of templates; as shown in FIGS. 4-5, the chatbot selects an appropriate scenario based on the natural-language dialogue exchanged with the user); [0186], The command generation processing unit 216 performs a command output control processing (for applying an algorithm to protected information responsive to the qualified request; according to [0085], the chatbot system may provide a variety of enterprise services. In the context of an enterprise environment, information relating to the organization’s structure, employee schedule, conference rooms and other office-related information may constitute protected information) that outputs the generated command to the selected server system 50 ... (Step S107)).
OHTA ‘109 does not disclose but Ponsini ‘028 discloses: validating the request in a first trusted computing environment (FIG. 1, [0049], Secure element 108 (a first trusted computing environment) also is configured for authentication and digital signature operations. Authentication ensures that access to online services is protected not only by a username and password, but by strong authentication based on credentials that are stored and processed inside secure element 108. Authentication (validating the request in the first computing environment) may be applied for logging-in to critical services, such as a virtual private network (VPN), corporate email, etc.).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified OHTA ‘109 with the secure element for authentication and digital signature operations to digital services as taught by Ponsini ‘028 because it would improve the security and resistance to software attacks by using isolated hardware execution and protected cryptographic assets [0044]. Additionally, Ponsini ‘028 is analogous to the claimed invention because it teaches a secure element (SE) hardware processor on a physical SE component located on a communication path between a peripheral component and a primary controller in a computing device [0032].
OHTA ‘109 in view of Ponsini ‘028 does not disclose but IMMS ‘865 discloses: when the user request is rejected, informing the user of the rejection (FIG. 1, [0034], Through the CLI 135 (where the user request is entered), the shell 130 allows a user to control the operations of the computer via entered keyboard commands; [0042], expand the capabilities of the shell 130 and the CU 135 by enabling the shell 135 and the CU 135 to have context awareness, as shown by context aware capabilities 145; FIG. 2, [0044], examples of context aware capabilities 200 (determining whether the user request is accepted or not). The above example was one type of example within the corrected version 205 subset of examples. The embodiments can analyze the command the user entered and determine whether the user meant to enter a different command. For instance, if the user mistyped a command (the user request is rejected), then the embodiments can propose a corrected version (e.g., a correctly typed version) of that command – informing the user of the rejection).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified OHTA ‘109 in view of Ponsini ‘028 with the context-aware command line interface (CLI) in which user-entered commands are analyzed to provide contextual actions or command suggestions as taught by IMMS ‘865 because the system improves human-computer interaction by enabling users to efficiently execute appropriate commands. Additionally, IMMS ‘865 is analogous to the claimed invention because it teaches that through the CLI 135, the shell 130 allows a user to control the operations of the computer via entered keyboard commands [0034].
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 does not disclose but Chalk ‘510 discloses: applying an algorithm to protected information responsive to the qualified request in a separate second trusted computing environment (FIG. 1A, [0041], The data steward uploads encrypted data to their cloud environment using an encrypted connection that terminates inside an Intel SGX-sequestered enclave (in a separate second trusted computing environment). Then, the algorithm developer submits an encrypted, containerized AI model which also terminates into an Intel SGX-sequestered enclave. A key management system in the management core enables the containers to authenticate (responsive to the qualified (or authenticated) request) and then run the model on the data (applying an algorithm to protected information) within the enclave. The data steward never sees the algorithm inside the container and the data is never visible to the algorithm developer).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 with the running of the model on the data in an Intel SGX-sequestered enclave by enabling the containers to authenticate as taught by Chalk ‘510 because this enables secure processing of sensitive datasets without exposing either party’s confidential assets. Additionally, Chalk ‘510 is analogous to the claimed invention because it teaches the data stewards maintain sequestered computing nodes 110a-y which function to actually perform the computation of the algorithm on the dataset [0039].
Per claim 2 (dependent on claim 1):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 discloses the elements detailed in the rejection of claim 1 above, incorporated herein by reference.
OHTA ‘109 discloses: The method of claim 1, further comprising receiving an output from the applying the algorithm to the protected information (FIG. 6, [0186], The command generation processing unit 216 performs a command output control processing (the applying the algorithm to the protected information) that outputs the generated command to the selected server system 50 ... (Step S107); [0187], Finally, when obtaining the information to enjoy the appropriate service (Step S108), command generation processing unit 216 provides the obtained information to the user terminal device 10 (receiving an output from the applying the algorithm) through the communication control unit 310 and terminates this operation (Step S109)).
Per claim 7 (dependent on claim 1):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 discloses the elements detailed in the rejection of claim 1 above, incorporated herein by reference.
OHTA ‘109 in view of Ponsini ‘028 and Chalk ‘510 does not disclose but IMMS ‘865 discloses: The method of claim 1, further comprising when the user request is rejected generating an amended request suggestion (FIG. 1, [0034], Through the CLI 135 (where the user request is entered), the shell 130 allows a user to control the operations of the computer via entered keyboard commands; [0042], expand the capabilities of the shell 130 and the CU 135 by enabling the shell 135 and the CU 135 to have context awareness, as shown by context aware capabilities 145; FIG. 2, [0044], examples of context aware capabilities 200 (determining whether the user request is rejected). The above example was one type of example within the corrected version 205 subset of examples. The embodiments can analyze the command the user entered and determine whether the user meant to enter a different command. For instance, if the user mistyped a command (when the user request is rejected), then the embodiments can propose a corrected version (e.g., a correctly typed version) of that command – generating an amended request suggestion).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified OHTA ‘109 in view of Ponsini ‘028 and Chalk ‘510 with the context-aware command line interface (CLI) in which user-entered commands are analyzed to provide contextual actions or command suggestions as taught by IMMS ‘865 because the system improves human-computer interaction by enabling users to efficiently execute appropriate commands.
Per claim 10 (dependent on claim 1):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 discloses the elements detailed in the rejection of claim 1 above, incorporated herein by reference.
OHTA ‘109 discloses: The method of claim 1, wherein the templates are in a format and content known to generate a favorable output from the algorithm (FIG. 2, [0119], In the scenario DB 332, two or more scenario data (the templates) is stored in association with IDs to identify each scenario data ... to allow the user to enjoy the each of the services; [0122], The "scenario" has a plurality of scripts to provide the service desired by a user, and defines the content and order of each scrips (that is, in a format), as well as a type of data required to execute the scripts and content of the data – to generate a favorable output from the algorithm).
Per claim 11 (independent):
The limitations of the claim(s) correspond(s) to features of claim 1 and the claim(s) is/are rejected for the reasons detailed with respect to claim 1.
Per claim 12 (dependent on claim 11):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 discloses the elements detailed in the rejection of claim 11 above, incorporated herein by reference.
Chalk ‘510 teaches “the second trusted computing environment”, and the remaining limitations of the claim(s) correspond(s) to features of claim 2 and the claim(s) is/are rejected for the reasons detailed with respect to claim 2.
Per claim 16 (dependent on claim 11):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 discloses the elements detailed in the rejection of claim 11 above, incorporated herein by reference.
Chalk ‘510 teaches “the second trusted computing environment”, and the remaining limitations of the claim(s) correspond(s) to features of claim 7 and the claim(s) is/are rejected for the reasons detailed with respect to claim 7.
Per claim 19 (dependent on claim 11):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 discloses the elements detailed in the rejection of claim 11 above, incorporated herein by reference.
The limitations of the claim(s) correspond(s) to features of claim 10 and the claim(s) is/are rejected for the reasons detailed with respect to claim 10.
Claim(s) 3, 8-9 and 17-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 and Blum et al., US-20250068667-A1 (hereinafter “Blum ‘667”).
Per claim 3 (dependent on claim 2):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 discloses the elements detailed in the rejection of claim 2 above, incorporated herein by reference.
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 does not disclose but Blum ‘667 discloses: The method of claim 2, further comprising validating the output to generate a final validation report (FIGS. 5-6, [0047], the response or completion 500 (the output) generated by the LLM 201 ... the response comprises a first part 510 (the output) that contains the response to the query 310. The response also comprises a second part 520 (by validating the output) that includes the verification data generated by the LLM 201 in response to the verification data request 320; FIG. 7, [0052], processing the verification data to determine whether the query result 510 comprises a hallucination; [0060], In step S703, ... If comparison reveals that the expression output and the corresponding value match, the value is valid. That is to say, if the expression output and value match, the value is not hallucinated. Conversely, if the expression output and value do not match, the value is determined to be hallucinated; [0066], The output of the comparisons (e.g. the verification table) may then be employed to label the query result 510. That is to say, having determined which values in the result 510 are valid and which are hallucinated, the occurrences of these values in the text of the result 510 may be tagged or labelled accordingly – generate a final validation report).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 with the verification of LLM-generated query results by evaluating verification data as taught by Blum ‘667 because hallucination content can be tagged or corrected before being presented to the user, thereby increasing the trustworthiness of the generated response. Additionally, Blum ‘667is analogous to the claimed invention because it teaches a method of processing the verification data to determine whether the query result 510 comprises a hallucination [0052].
Per claim 8 (dependent on claim 1):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 discloses the elements detailed in the rejection of claim 1 above, incorporated herein by reference.
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 does not disclose but Blum ‘667 discloses: The method of claim 1, wherein the matching includes applying a Large Language Model (LLM) to the user request (FIG. 2, [0029], a process of constructing a prompt 300 for submission to the LLM 201 (LLM); [0030], The prompt 300 (the user request applied to the LLM) comprises a first part 310, which includes query text 311 and input data 312. The query text 311 prompts the LLM 201 to provide a query result, wherein the result is derived from the input data 312; [0031], The query text 311 may be predetermined text, stored in storage 120. For example, the system 100 may select (the matching) amongst one of a plurality of predetermined query texts 122 depending upon circumstance or in response to a particular user input).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 with the submission of a prompt to a LLM for verification of LLM-generated query results based on the user request by evaluating verification data as taught by Blum ‘667 because hallucination content can be tagged or corrected before being presented to the user, thereby increasing the trustworthiness of the generated response.
Per claim 9 (dependent on claim 8):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 and Blum ‘667 discloses the elements detailed in the rejection of claim 8 above, incorporated herein by reference.
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 does not disclose but Blum ‘667 discloses: The method of claim 8, wherein the LLM extracts context and substantive data from the user request (FIG. 2, [0029], a process of constructing a prompt 300 for submission to the LLM 201 (LLM); [0030], The prompt 300 (the user request applied to the LLM) comprises a first part 310, which includes query text 311 and input data 312. The query text 311 prompts the LLM 201 to provide a query result, wherein the result is derived from the input data 312; FIG. 5, [0047], the response or completion 500 generated by the LLM 201 ... the response comprises a first part 510 (based on context and substantive data from the user request included in the prompt 300) that contains the response to the query 310).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 with the submission of a prompt to a LLM for verification of LLM-generated query results based on the user request by evaluating verification data as taught by Blum ‘667 because hallucination content can be tagged or corrected before being presented to the user, thereby increasing the trustworthiness of the generated response.
Per claim 17 (dependent on claim 11):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 discloses the elements detailed in the rejection of claim 11 above, incorporated herein by reference.
The limitations of the claim(s) correspond(s) to features of claim 8 and the claim(s) is/are rejected for the reasons detailed with respect to claim 8.
Per claim 18 (dependent on claim 17):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 and Blum ‘667 discloses the elements detailed in the rejection of claim 12 above, incorporated herein by reference.
The limitations of the claim(s) correspond(s) to features of claim 9 and the claim(s) is/are rejected for the reasons detailed with respect to claim 9.
Claim(s) 4 is/are rejected under 35 U.S.C. 103 as being unpatentable over OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 and Blum ‘667 and Vahldiek-Oberwagner et al., US-20240305465-A1 (hereinafter “Vahldiek ‘465”).
Per claim 4 (dependent on claim 3):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 and Blum ‘667 discloses the elements detailed in the rejection of claim 3 above, incorporated herein by reference.
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 and Blum ‘667 does not disclose but Vahldiek ‘465 discloses: The method of claim 3, wherein the validating the output to generate the final validation report is performed in a third trusted computing environment (FIG. 1, [0019], example accuracy evaluation circuitry 100 of an example artificial intelligence (AI) model 102 ... operates in an example TEE 104 (in a third trusted computing environment); [0022], The signed artifact 130 (the final validation report) is a data packet that includes information that can be used to verify how the accuracy of the AI model 102 was evaluated and how the accuracy value was obtained during the process; [0027], the platform and/or enclave meta 208 includes information related to the hardware used in the TEE (the third trusted computing environment) by the trusted third party to operate the software applied by the trusted third party 108; note that the input data set 110 may serve as validation data for evaluating the AI model, and the resulting accuracy statistics 204 generated from the validation data may be incorporated into the signed artifact 130).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 and Blum ‘667 with the evaluation of the AI model using validation data within a trusted execution environment as taught by Vahldiek ‘465 because the system ensures that the reported accuracy statistics have not been tampered with, thereby enabling third parties to independently verify the integrity of the model. Additionally, Vahldiek ‘465 is analogous to the claimed invention because it teaches example accuracy evaluation circuitry 100 of an example artificial intelligence (AI) model 102 [0019].
Claim(s) 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 and Vahldiek ‘465.
Per claim 13 (dependent on claim 12):
OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 discloses the elements detailed in the rejection of claim 12 above, incorporated herein by reference.
The limitations of the claim(s) correspond(s) to features of claim 4 and the claim(s) is/are rejected for the reasons detailed with respect to claim 4.
Allowable Subject Matter
Claim(s) 5-6 and 14-15 is/are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
The following is a statement of reasons for the indication of allowable subject matter:
Regarding claim 5 (and 14), the prior art of record (OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 and Vahldiek ‘465) does not disclose: “the first trusted computing environment, second trusted computing environment, and third trusted computing environment all exist within the computing infrastructure of a data steward.”
Although Ponsini ‘028, Chalk ‘510, and Vahldiek ‘465 may be relied upon as teaching the first, second, and third trusted computing environments, respectively, none of the cited references teaches or suggests that these three distinct trusted environments are all present within a data steward, i.e., a computing device.
Regarding claim 6 (and 15), the prior art of record (OHTA ‘109 in view of Ponsini ‘028 and IMMS ‘865 and Chalk ‘510 and Vahldiek ‘465) does not disclose: “the validating the request includes approving permissions, approving request format, approving output format, approving background information on objectives of the user request, approving background information on subject matter of the user request, approving a security level, approving a processing model and approving the data source restrictions.”
While the above combination of references teaches “validating the request” by, for example, “approving permissions and approving output format,” none of the cited combinations teaches or suggests satisfying all of the above-recited conditions required for the validation.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Ferenczi, US-20250088365-A1 – this illustrates a chat authentication in which a post-quantum cryptographically signed chat request is verified before an utterance is forwarded to a chatbot service. Upon successful signature validation, the request is approved, the utterance history may be updated, and the chatbot response is returned to the client device.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SANGSEOK PARK whose telephone number is (571)272-4332. The examiner can normally be reached Monday-Friday 7:30-5:30 and Alternate Fridays 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, PHILIP CHEA can be reached at (571)272-3951. 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.
/SANGSEOK PARK/Primary Examiner, Art Unit 2499
Prosecution Insights