DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This office action is in response to the application filed March 2, 2026.
Claims 1-20 are pending.
Priority
Applicant’s claim for the benefit of prior-filed applications under 35 U.S.C. 119(e) or under 35 U.S.C. 120, 121, 365(c), or 386(c) is acknowledged. Applicant has not complied with one or more conditions for receiving the benefit of an earlier filing date under 35 U.S.C. §112 as follows: The later-filed application must be an application for a patent for an invention which is also disclosed in the prior application (the parent or original nonprovisional application or provisional application). The disclosure of the invention in the parent applications and in the later-filed application must be sufficient to comply with the requirements of 35 U.S.C. 112(a) or the first paragraph of pre-AIA 35 U.S.C. 112, except for the best mode requirement. See Transco Products, Inc. v. Performance Contracting, Inc., 38 F.3d 551, 32 USPQ2d 1077 (Fed. Cir. 1994).
The disclosure of the prior-filed applications, Application No. 19/279103 19/204706 18/951120 18/907414 18/830573 18/812913 18/661519 18/661532 18/633293 fail to provide adequate support or enablement in the manner provided by 35 U.S.C. 112(a) or pre-AIA 35 U.S.C. 112, first paragraph for claims 1-20 as pending.
The claims herein are examined given the filing date of the current application September 12, 2025.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on March 2, 2026 was filed after the mailing date of the application on September 12, 2025 The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Response to Arguments
Applicant's arguments filed March 2, 2026 have been fully considered but they are not persuasive.
Specifically, Applicant’s remarks with respect to the amended limtaitons and the teachings of Yeo and Agardh are not persuasive. Specifically, while applicant argue Yeo’s system including buckets for the tokens for various actions does not teach “token types corresponding to different computational resource types such as processing power, memory, storage, or bandwidth…” (Remarks, Page 9) even applicant’s amended claim does not require such a teaching. Specifically, while applicant amended to require “each token type of the corresponding token types corresponds to a respective computational resource type of the set of computational resources” the amended language does not limit the “computation resource type” to the categories argued. Specifically Yeo teaches the different buckets of token correspond to different computation usage actions for file systems or tenants. The Examiner appreciates this may be different than embodiments disclosed in applicant’s invention, but the claim as amended is still taught or suggested by this teaching of Yeo as the claim does not require, for example, bandwidth token types, power token types, memory token types etc, but instead simply that the correspond to computation resources. As such, no such limitation will be read into the claims where applicant has not explicitly chosen to amend in such a manner. Moreover, Applicant’s arguments and amendments regarding the recording of an identifier in the distributed ledger is also not persuasive. Specifically, the amendment does not specify a type of identifier such that it would put any limit on the type of identification and, for example, Arardh teaches using an service description parameter for identifying the type of resource usage recorded in the ledger in e.g. ¶¶59-60 as one example of identifier recordation that would be well known in the art. As such, these arguments are unpersuasive and the rejection is maintained.
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) 1,5,7,11,13,14,18 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over “Elbadrashiny” (US Patent 12,307,299) in view of “Yeo” (US Patent 12,174,793) and further in view of “Agardh” (US PG Pub 2022/0141154).
Regarding Claim 1, Elbadrashiny teaches:
1. One or more non-transitory, computer-readable storage media comprising instructions recorded thereon for facilitating access to computational resources by agents, wherein the instructions, when executed by at least one data processor of a system, (Elbadrashiny 106, Fig. 1B-E,Fig.3 and Fig. 4, e.g. Col.8 Ln12-Col. 9, Further see Col. 10, Ln 21-33, teaches an AI agent management system which executes the tasks of AI agents on system hardware resources including processors, memory and storage elements)
cause the system to: determine that a set of computational resources is available for consumption by a plurality of artificial intelligence (Al) agents, the set of computational resources comprising one or more of processing power, memory, storage, or bandwidth, (Elbadrashiny 106, Fig. 1B-E,Fig.3 and Fig. 4, e.g. Col.8 Ln12-Col. 9, Further see Col. 10, Ln 21-33, teaches an AI agent management system which executes the tasks of AI agents on system hardware resources including processors, memory and storage elements) [Here, as written applicant’s claim requires one of the listed types of resources, Elbadrashiny however teaches or suggests at least processing power, memory and storage elements in its AI agent execution system]
wherein one or more of the plurality of Al agents is enabled to access computational resources (Elbadrashiny 106, Fig. 1B-E,Fig.3 and Fig. 4, e.g. Col.8 Ln12-Col. 9, Further see Col. 10, Ln 21-33, teaches an AI agent management system which executes the tasks of AI agents on system hardware resources including processors, memory and storage elements, See further Fig. 3, Col. 16, Ln)
and wherein the [AI agents] are associated with a corresponding computer-executable operation set configured to be autonomously executed by a corresponding Al agent on a software application set; (Elbadrashiny 106, Fig. 1B-E,Fig.3 and Fig. 4, e.g. Col.8 Ln12-Col. 9, Further see Col. 10, Ln 21-33, teaches an AI agent management system which executes the tasks of AI agents on system hardware resources including processors, memory and storage elements, See further Fig. 3, Col. 16, Ln)
Elbadrashiny does not explicitly teach, but Yeo teaches:
by exchanging tokens for access to the computational resources, (See Yeo Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 – teaches populating buckets of tokens for tenant processes based on system computational capacity, and teaches exchanging tokens in proportion to the priority for computational resources)
generate a set of tokens of corresponding token types for the set of computational resources wherein each token type of the corresponding token types corresponds to a respective computational resource type of the set of computational resources; (See Yeo Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 above, see further 206, 208 Fig. 2 and Col. 6, Ln27-44 teaches that token buckets can be filled in relation to system capacity and may be periodically refilled) [Here applicant’s claim does not limit the character of “token types” claimed, and as such, token types claimed here are interpreted as anticipated by the different buckets of tokens set forth in Yeo which each provide access to computing resources in Yeo’s system. Further, Applicant’s amended language states that the token types correspond to a computational resource type, but fails to outline what specifically those computation resource types are, nor how they correspond. As such, this amended limitation is taught by the system in Yeo which for the buckets correspond to computational resource usage for tenants and file systems respectively. Should applicant desire to limit the claim to computational resource types specifically defined as for example a a memory token type, a storage token type, a bandwidth token type etc such language would be needed in the claim amendment. No such limitation will be read into the claim absence specific limiting language]
receive a set of requests, from a set of [AI agents] of the plurality of [AI agents], to access a first computational resource of the set of computational resources, each request comprising a number of priority tokens associated with the request; ((See Yeo Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 – teaches populating buckets of tokens for tenant processes based on system computational capacity, and teaches exchanging tokens in proportion to the priority for computational resources, Yeo Fig. 2 and Fig. 4, Col. 15, Ln 50 to Col. 16 Ln 47 teaches exchange of tokens from buckets for resource usage for task execution in Yeo’s system) [Here, while Yeo does not teach the use of AI agents, such agents are taught above by Elbadrashiny and both systems teach the allocation of tasks for execution in computing systems and Yeo teaches the use of priority-indicating tokens in exchange for system resource usage]
determine a queue of the set of requests, wherein an order of the queue is based on the number of priority tokens associated with each request; (Yeo e.g. Col. 6, Ln 65 to Col. 7, Ln 47 and Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 – teaches populating buckets of tokens for tenant processes based on system computational capacity, and teaches exchanging tokens in proportion to the priority for computational resources, and forming queues for the resources based on the requests and tokens. Yeo Fig. 2 and Fig. 4, Col. 15, Ln 50 to Col. 16 Ln 47 teaches exchange of tokens from buckets for resource usage for task execution in Yeo’s system)
perform, based on the order of the queue, a transfer of one or more tokens of a first type corresponding to the first computational resource to one or more [AI agents] of the set of [AI agents] in exchange for the number of priority tokens associated with each respective request, wherein the one or more [AI agents] are enabled to gain access to the first computational resource by exchanging the one or more tokens for the first computational resource; ((See Yeo Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 – teaches populating buckets of tokens for tenant processes based on system computational capacity, and teaches exchanging tokens in proportion to the priority for computational resources, Yeo Fig. 2 and Fig. 4, Col. 15, Ln 50 to Col. 16 Ln 47 teaches exchange of tokens from buckets for resource usage for task execution in Yeo’s system) [Here, while Yeo does not teach the use of AI agents, such agents are taught above by Elbadrashiny and both systems teach the allocation of tasks for execution in computing systems and Yeo teaches the use of priority-indicating tokens in exchange for system resource usage]
and record, …the one or more tokens transferred to the one or more [AI agents], and the number of priority tokens exchanged for the first computational resource. (Yeo 122,124, Fig. 1 Col. 7, Ln 27 to Col. 8, Ln 24. Inherent in this part of Yeo is that there must necessiarly be some recording of the number of tokens assigned to tenant buckets, and the decrementing of those tokens in each transaction as seen in Fig 4, 408 and 412. Recording such token usage would be necessary to do the token bucket tracking and interaction described in Yeo) [Here, while Yeo does not teach the recording of the records of the token’s in a distributed ledger, use of such a ledger would be obvious to one of ordinary skill as further described herein below]
In addition, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the application to apply the teaching of Yeo to the system of Elbadrashiny as each is directed to computation resource management systems and Yeo recognized a need that “requests such that the file system environment may experience latency in performing the requested actions, may fail to perform the requested actions within requested parameters, or may experience other technical problems” (Yeo Col. 1 Ln14-19) and provided a prioritization system that remedies that issue.
Elbadrashiny further does not teach, but Agardh teaches:
via a distributed ledger, an identifier of each of the one or more [AI agents] involved in the transfer, (See Agardh Fig. 11, 106-108, e.g. ¶¶53,59-60,186 teaches recording resource usage for computing resources in an execution environment on a distributed ledger) [While Agardh does not teach AI agents per se, such agents are taught by Elbadrashiny as described above]
In addition, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the application to apply the teaching of Agardh to the system of Elbadrashiny as each is directed to computation resource management systems and Agardh recognized their system “may be particularly suitable when the resource allocation is conducted in terms of leasing of resources, e.g. as time allocation and/or location-specific allocation and/or service-specific allocation of a resource from the resource allocation servers. The leasing of spectrum can in such case be considered as a transaction, for example where storing parameters defining agreements in a transaction can be made via a resource allocation distributed ledger functionality.” (Agardh ¶53).
Regarding Claim 7, Elbadrashiny teaches:
7. A method comprising: determining that a set of computational resources is available for consumption by a plurality of models, wherein the plurality of models is enabled to access computational resources (Elbadrashiny 106, Fig. 1B-E,Fig.3 and Fig. 4, e.g. Col.8 Ln12-Col. 9, Further see Col. 10, Ln 21-33, teaches an AI agent management system which executes the tasks of AI agents on system hardware resources including processors, memory and storage elements; see further Col. 2, Ln 4-30; Col. 5, Ln 27-66 describing agent’s implemented as LL models)
Elbadrashiny does not explicitly teach, but Yeo teaches:
by exchanging tokens for the computational resources; (See Yeo Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 – teaches populating buckets of tokens for tenant processes based on system computational capacity, and teaches exchanging tokens in proportion to the priority for computational resources)
generating a set of tokens of corresponding token types for the set of computational resources wherein each token type of the corresponding token types corresponds to a respective computational resource type of the set of computational resources;; (See Yeo Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 above, see further 206, 208 Fig. 2 and Col. 6, Ln27-44 teaches that token buckets can be filled in relation to system capacity and may be periodically refilled) [Here applicant’s claim does not limit the character of “token types” claimed, and as such, token types claimed here are interpreted as anticipated by the different buckets of tokens set forth in Yeo which each provide access to computing resources in Yeo’s system. Further, Applicant’s amended language states that the token types correspond to a computational resource type, but fails to outline what specifically those computation resource types are, nor how they correspond. As such, this amended limitation is taught by the system in Yeo which for the buckets correspond to computational resource usage for tenants and file systems respectively. Should applicant desire to limit the claim to computational resource types specifically defined as for example a a memory token type, a storage token type, a bandwidth token type etc such language would be needed in the claim amendment. No such limitation will be read into the claim absence specific limiting language]
receiving a set of requests, from a set of [models] of the plurality of [models], to access a first computational resource of the set of computational resources, each request comprising a number of priority tokens associated with the request; ((See Yeo Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 – teaches populating buckets of tokens for tenant processes based on system computational capacity, and teaches exchanging tokens in proportion to the priority for computational resources, Yeo Fig. 2 and Fig. 4, Col. 15, Ln 50 to Col. 16 Ln 47 teaches exchange of tokens from buckets for resource usage for task execution in Yeo’s system) [Here, while Yeo does not teach the use of AI agents, such agents are taught above by Elbadrashiny and both systems teach the allocation of tasks for execution in computing systems and Yeo teaches the use of priority-indicating tokens in exchange for system resource usage]
determining a queue of the set of requests, wherein an order of the queue is based on the number of priority tokens associated with each request; (Yeo e.g. Col. 6, Ln 65 to Col. 7, Ln 47 and Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 – teaches populating buckets of tokens for tenant processes based on system computational capacity, and teaches exchanging tokens in proportion to the priority for computational resources, and forming queues for the resources based on the requests and tokens. Yeo Fig. 2 and Fig. 4, Col. 15, Ln 50 to Col. 16 Ln 47 teaches exchange of tokens from buckets for resource usage for task execution in Yeo’s system)
performing, based on the order of the queue, a transfer of one or more tokens of a first type corresponding to the first computational resource to one or more [models] of the set of [models] in exchange for the number of priority tokens associated with each respective request, wherein the one or more [models] are enabled to gain access to the first computational resource by exchanging the one or more tokens for the first computational resource; ((See Yeo Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 – teaches populating buckets of tokens for tenant processes based on system computational capacity, and teaches exchanging tokens in proportion to the priority for computational resources, Yeo Fig. 2 and Fig. 4, Col. 15, Ln 50 to Col. 16 Ln 47 teaches exchange of tokens from buckets for resource usage for task execution in Yeo’s system) [Here, while Yeo does not teach the use of AI agents, such agents are taught above by Elbadrashiny and both systems teach the allocation of tasks for execution in computing systems and Yeo teaches the use of priority-indicating tokens in exchange for system resource usage]
[record]…the one or more tokens transferred to the one or more [models], and the number of priority tokens exchanged for the first computational resource. (Yeo 122,124, Fig. 1 Col. 7, Ln 27 to Col. 8, Ln 24. Inherent in this part of Yeo is that there must necessiarly be some recording of the number of tokens assigned to tenant buckets, and the decrementing of those tokens in each transaction as seen in Fig 4, 408 and 412. Recording such token usage would be necessary to do the token bucket tracking and interaction described in Yeo) [Here, while Yeo does not teach the recording of the records of the token’s in a distributed ledger, use of such a ledger would be obvious to one of ordinary skill as further described herein below]
In addition, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the application to apply the teaching of Yeo to the system of Elbadrashiny as each is directed to computation resource management systems and Yeo recognized a need that “requests such that the file system environment may experience latency in performing the requested actions, may fail to perform the requested actions within requested parameters, or may experience other technical problems” (Yeo Col. 1 Ln14-19) and provided a prioritization system that remedies that issue.
Elbadrashiny further does not teach, but Agardh teaches:
and recording, via a distributed ledger, an identifier of each of the one or more [models] involved in the transfer, (See Agardh Fig. 11, 106-108, e.g. ¶¶53,59-60,186 teaches recording resource usage for computing resources in an execution environment on a distributed ledger) [While Agardh does not teach AI agents per se, such agents are taught by Elbadrashiny as described above]
In addition, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the application to apply the teaching of Agardh to the system of Elbadrashiny as each is directed to computation resource management systems and Agardh recognized their system “may be particularly suitable when the resource allocation is conducted in terms of leasing of resources, e.g. as time allocation and/or location-specific allocation and/or service-specific allocation of a resource from the resource allocation servers. The leasing of spectrum can in such case be considered as a transaction, for example where storing parameters defining agreements in a transaction can be made via a resource allocation distributed ledger functionality.” (Agardh ¶53).
Regarding Claim 14, Elbadrashiny teaches:
14. A system comprising:a storage device; (e.g. 702, Fig. 7) and one or more processors (e.g. 718, Fig. 7) communicatively coupled to the storage device storing instructions thereon, that cause the one or more processors to: determine that a set of computational resources is available for consumption by a plurality of models, (Elbadrashiny 106, Fig. 1B-E,Fig.3 and Fig. 4, e.g. Col.8 Ln12-Col. 9, Further see Col. 10, Ln 21-33, teaches an AI agent management system which executes the tasks of AI agents on system hardware resources including processors, memory and storage elements; see further Col. 2, Ln 4-30; Col. 5, Ln 27-66 describing agent’s implemented as LL models)
wherein the plurality of models is enabled to access computational resources (Elbadrashiny 106, Fig. 1B-E,Fig.3 and Fig. 4, e.g. Col.8 Ln12-Col. 9, Further see Col. 10, Ln 21-33, teaches an AI agent management system which executes the tasks of AI agents on system hardware resources including processors, memory and storage elements; see further Col. 2, Ln 4-30; Col. 5, Ln 27-66 describing agent’s implemented as LL models)
Elbadrashiny does not explicitly teach, but Yeo teaches:
by exchanging tokens for the computational resources; (See Yeo Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 – teaches populating buckets of tokens for tenant processes based on system computational capacity, and teaches exchanging tokens in proportion to the priority for computational resources)
generate a set of tokens of corresponding token types for the set of computational resources wherein each token type of the corresponding token types corresponds to a respective computational resource type of the set of computational resources;; (See Yeo Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 above, see further 206, 208 Fig. 2 and Col. 6, Ln27-44 teaches that token buckets can be filled in relation to system capacity and may be periodically refilled) [Here applicant’s claim does not limit the character of “token types” claimed, and as such, token types claimed here are interpreted as anticipated by the different buckets of tokens set forth in Yeo which each provide access to computing resources in Yeo’s system. Further, Applicant’s amended language states that the token types correspond to a computational resource type, but fails to outline what specifically those computation resource types are, nor how they correspond. As such, this amended limitation is taught by the system in Yeo which for the buckets correspond to computational resource usage for tenants and file systems respectively. Should applicant desire to limit the claim to computational resource types specifically defined as for example a a memory token type, a storage token type, a bandwidth token type etc such language would be needed in the claim amendment. No such limitation will be read into the claim absence specific limiting language]
receive a set of requests, from a set of [models] of the plurality of [models], to access a first computational resource of the set of computational resources, each request comprising a number of priority tokens associated with the request; ((See Yeo Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 – teaches populating buckets of tokens for tenant processes based on system computational capacity, and teaches exchanging tokens in proportion to the priority for computational resources, Yeo Fig. 2 and Fig. 4, Col. 15, Ln 50 to Col. 16 Ln 47 teaches exchange of tokens from buckets for resource usage for task execution in Yeo’s system) [Here, while Yeo does not teach the use of AI agents, such agents are taught above by Elbadrashiny and both systems teach the allocation of tasks for execution in computing systems and Yeo teaches the use of priority-indicating tokens in exchange for system resource usage]
determine a queue of the set of requests, wherein an order of the queue is based on the number of priority tokens associated with each request; (Yeo e.g. Col. 6, Ln 65 to Col. 7, Ln 47 and Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 – teaches populating buckets of tokens for tenant processes based on system computational capacity, and teaches exchanging tokens in proportion to the priority for computational resources, and forming queues for the resources based on the requests and tokens. Yeo Fig. 2 and Fig. 4, Col. 15, Ln 50 to Col. 16 Ln 47 teaches exchange of tokens from buckets for resource usage for task execution in Yeo’s system)
perform, based on the order of the queue, a transfer of one or more tokens of a first type corresponding to the first computational resource to one or more [models] of the set of [models] in exchange for the number of priority tokens associated with each respective request, wherein the one or more [models] are enabled to gain access to the first computational resource by exchanging the one or more tokens for the first computational resource; ((See Yeo Fig. 1, 2, 4, Col. 7, Ln 47 to Col. 8, Ln 24 – teaches populating buckets of tokens for tenant processes based on system computational capacity, and teaches exchanging tokens in proportion to the priority for computational resources, Yeo Fig. 2 and Fig. 4, Col. 15, Ln 50 to Col. 16 Ln 47 teaches exchange of tokens from buckets for resource usage for task execution in Yeo’s system) [Here, while Yeo does not teach the use of AI agents, such agents are taught above by Elbadrashiny and both systems teach the allocation of tasks for execution in computing systems and Yeo teaches the use of priority-indicating tokens in exchange for system resource usage]
record…the one or more tokens transferred to the one or more [models], and the number of priority tokens exchanged for the first computational resource. (Yeo 122,124, Fig. 1 Col. 7, Ln 27 to Col. 8, Ln 24. Inherent in this part of Yeo is that there must necessarily be some recording of the number of tokens assigned to tenant buckets, and the decrementing of those tokens in each transaction as seen in Fig 4, 408 and 412. Recording such token usage would be necessary to do the token bucket tracking and interaction described in Yeo) [Here, while Yeo does not teach the recording of the records of the token’s in a distributed ledger, use of such a ledger would be obvious to one of ordinary skill as further described herein below]
In addition, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the application to apply the teaching of Yeo to the system of Elbadrashiny as each is directed to computation resource management systems and Yeo recognized a need that “requests such that the file system environment may experience latency in performing the requested actions, may fail to perform the requested actions within requested parameters, or may experience other technical problems” (Yeo Col. 1 Ln14-19) and provided a prioritization system that remedies that issue.
Elbadrashiny further does not teach, but Agardh teaches:
and record, via a distributed ledger, an identifier of each of the one or more models involved in the transfer, (See Agardh Fig. 11, 106-108, e.g. ¶¶53,59-60,186 teaches recording resource usage for computing resources in an execution environment on a distributed ledger) [While Agardh does not teach models per se, such agents as models are taught by Elbadrashiny as described above]
In addition, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the application to apply the teaching of Agardh to the system of Elbadrashiny as each is directed to computation resource management systems and Agardh recognized their system “may be particularly suitable when the resource allocation is conducted in terms of leasing of resources, e.g. as time allocation and/or location-specific allocation and/or service-specific allocation of a resource from the resource allocation servers. The leasing of spectrum can in such case be considered as a transaction, for example where storing parameters defining agreements in a transaction can be made via a resource allocation distributed ledger functionality.” (Agardh ¶53).
Regarding Claim 5, Elbadrashiny does not teach, but Yeo further teaches:
5. The one or more non-transitory, computer-readable storage media of claim 1, wherein the instructions further cause the system to remove tokens of a particular type from circulation within the system based on an agent included in the plurality of Al agents consuming a respective computational resource corresponding to the token. (Yeo 122,124, Fig. 1 Col. 7, Ln 27 to Col. 8, Ln 24. Inherent in this part of Yeo is that there must necessiarly be some recording of the number of tokens assigned to tenant buckets, and the decrementing of those tokens in each transaction as seen in Fig 4, 408 and 412. Recording such token usage would be necessary to do the token bucket tracking and interaction described in Yeo) In addition, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the application to apply the teaching of Yeo to the system of Elbadrashiny as each is directed to computation resource management systems and Yeo recognized a need that “requests such that the file system environment may experience latency in performing the requested actions, may fail to perform the requested actions within requested parameters, or may experience other technical problems” (Yeo Col. 1 Ln14-19) and provided a prioritization system that remedies that issue.
Claims 11 and 18 are rejected on the same basis as claim 5 above.
Regarding Claim 13, Elbadrashiny teaches:
13. The method of claim 7, wherein the set of computational resources comprises one or more of processing power, memory, storage, or bandwidth. (Elbadrashiny 106, Fig. 1B-E,Fig.3 and Fig. 4, e.g. Col.8 Ln12-Col. 9, Further see Col. 10, Ln 21-33, teaches an AI agent management system which executes the tasks of AI agents on system hardware resources including processors, memory and storage elements) [Here, as written applicant’s claim requires one of the listed types of resources, Elbadrashiny however teaches or suggests at least processing power, memory and storage elements in its AI agent execution system]
Claim 20 is rejected on the same basis as claim 13 above.
Claim(s) 2, 8, and 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over “Elbadrashiny” (US Patent 12,307,299) in view of “Yeo” (US Patent 12,174,793) and further in view of “Agardh” (US PG Pub 2022/0141154) as applied above and further in view of “Terakawa” (US PG Pub 2008/0112313).
Regarding Claim 2, Elbadrashiny et al teach the limitations of claim 1 above but do not further teach, but Terkawa teaches:
2. The one or more non-transitory, computer-readable storage media of claim 1, wherein the instructions further cause the system to enable a first agent included in the plurality of Al agents and having idle computational resources to share the idle computational resources with a second agent included in the plurality of Al agents in exchange for corresponding tokens. (See Terakawa e.g. ¶¶26-28 teaches transferring unused tokens for resources from one requester group to another for execution)
In addition, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the application to apply the teaching of Terakawa to the system of Elbadrashiny as each is directed to computation resource management systems and recognized a need that “certain requesters may have to wait for a resource. This is wasteful if another group has one or more available tokens to the resource that are unused.” (¶5).
Claims 8 and 15 are rejected on the same basis as claim 2 above.
Claim(s) 3,9, and16 is/are rejected under 35 U.S.C. 103 as being unpatentable over “Elbadrashiny” (US Patent 12,307,299) in view of “Yeo” (US Patent 12,174,793) and further in view of “Agardh” (US PG Pub 2022/0141154) as applied above and further in view of “Kirk” (US PG Pub 2025/0123894).
Regarding Claim 3, Elbadrashiny et al teach the limitations of claim 1 above but do not further teach, but Kirk teaches:
3. The one or more non-transitory, computer-readable storage media of claim 1, wherein the instructions further cause the system to update, in real time, a cost of each token type for the set of computational resources based on at least one of resource utilization, transaction volume, historical pricing, or external demand factors. (Kirk ¶¶30,37,42 teaches a resource allocation system for AI-bots that updates utilization and cost constraints in realtime)
In addition, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the application to apply the teaching of Kirk to the system of Elbadrashiny as each is directed to computation resource management systems and Kirk recognized a need that “the dynamic nature of compute processes and their constraints may require a system to adapt to conditions in real-time.” (Kirk ¶15).
Claims 9 and 16 are rejected on the same basis as claim 3 above.
Claim(s) 4,10, and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over “Elbadrashiny” (US Patent 12,307,299) in view of “Yeo” (US Patent 12,174,793) and further in view of “Agardh” (US PG Pub 2022/0141154) as applied above and further in view of “Randhi” (US PG Pub 2024/0345890).
Regarding Claim 4, Elbadrashiny et al teach the limitations of claim 1 above and further teaches: 4. The one or more non-transitory, computer-readable storage media of claim 1, wherein the instructions further cause the system to that automatically adjust allocation of computational resources among the plurality of Al agents based on real-time system metrics, the real-time system metrics comprising current system load, historical usage patterns, and predictive models of demand. (Elbadrashiny see Col. 13, Ln 1-41 teaches using historical data and current workload as part of the process for improving allocations, as well as adjusting future decisions related to demand. See further e.g. Col. 5, Ln 29-45 regarding predicting future purchasing via modeling).
Elbadrashiny does not teach, but Randhi teaches:
implement smart contracts (See Randhi e.g. ¶69 teaching the use of smart contracts for real-time optimization of a workload management platform) In addition, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the application to apply the teaching of Randhi to the system of Elbadrashiny as each is directed to computation resource management systems and Randhi provides a system where “configuration output may be validated based on achieving consensus approval from a plurality of contract approvers” (Randhi ¶9).
Claims 10 and 17 are rejected on the same basis as claim 4 above.
Claim(s) 6, 12 and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over “Elbadrashiny” (US Patent 12,307,299) in view of “Yeo” (US Patent 12,174,793) and further in view of “Agardh” (US PG Pub 2022/0141154) as applied above and further in view of “LLMChain” (Bouchiha, Mouhamed Amine, et al. "Llmchain: Blockchain-based reputation system for sharing and evaluating large language models." 2024 IEEE 48th Annual Computers, Software, and Applications Conference (COMPSAC). IEEE, 2024.).
Regarding Claim 6, Elbadrashiny et al teach the limitations of claim 1 above but do not further teach, but LLMChain teaches:
6. The one or more non-transitory, computer-readable storage media of claim 1, wherein the instructions further cause the system to maintain a reputation system configured to track reputation scores for models included in the plurality of Al agents based on reliability, performance data, and disputes involving the models included in the plurality of Al agents, and wherein determining access to shared computational resources by the models included in the plurality of Al agents is based at least in part on the reputation scores.(LLMChain e.g. Section III teaches “a decentralized reputation-based store that allows sharing and evaluating LLMs. It serves a
dual role by addressing the needs of users seeking reliable AI assistance, as well as assisting LLMs developers in enhancing the performance and reliability of their models. Fig.1 illustrates an overview of the proposed LLMChain framework.”) [Inherent here is that LLMChain’s reputation based system may be used in selecting models for applications such as the AI agents in Elbadrashiny]
In addition, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the application to apply the teaching of LLMChain to the system of Elbadrashiny as each is directed to model deployment systems and LLMChain teaches a system which tackles issues that “present hurdles to the utilization of LLMs in critical contexts such as medical diagnosis, legal advice, or sensitive information processing, where accuracy and reliability are essential.” (LLMChain Section I.).
Claims 12 and 19 are rejected on the same basis as claim 3 above.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. The attached PTO-892 form includes additional references relevant to the disclosures in the application related to AI agent deployment and resource allocation systems.
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 MATTHEW J BROPHY whose telephone number is (571)270-1642. The examiner can normally be reached Monday-Friday, 9am-4:30pm.
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, Wei Zhen can be reached at 571-272-3708. 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.
MJB
4/17/2026
/MATTHEW J BROPHY/Primary Examiner, Art Unit 2191
Prosecution Insights