DETAILED ACTION
1. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . The following FINAL office action is in response to Applicant communication filed on 08/04/2026 regarding application 18/671,311. Claims 1, 11 and 20 has been amended. Claims 3 and 13 have been canceled. Claims 23-24 have been added as new claims. Claims 1, 4-11 and 14-24 are pending and have been rejected.
Response to Amendments
2. Applicant’s amendment filed on 08/04/2026 necessitated new grounds of rejection in this office action.
Priority
3. The Examiner has noted the Applicants claiming Priority from Continuation in Part (CIP) of Application #18/474,492 filed on 09/26/2023.
IDS Statements
4. The 1 Information Disclosure Statement (IDS) filed on 06/03/2026 complies with the provisions of 37 CFR 1.97, 1.98 and MPEP § 609 and is considered by the Examiner.
Response to 35 U.S.C. § 101 Arguments
5. Applicant’s 35 U.S.C. § 101 arguments, filed with respect to Claims 1, 4-11 and 14-24 have been fully considered, but they are found not persuasive (see Applicant Remarks, Pages 1-3 dated 08/04/2026). Examiner respectfully disagrees.
Argument #1:
(A). Applicant argues that Claims 1, 4-11 and 14-24 recite additional elements that integrate the judicial exception into a practical application under revised step 2a prong two of the 35 U.S.C. § 101 analysis (see Applicant Remarks, Page 2, dated 08/04/2026). Examiner respectfully disagrees.
Specifically, Applicant argues the amended claims 1, 11 and 20 which recite "each machine learning model having a limited size and including curated training data specific to the different combination of the plurality of subsystems" and "wherein the limited size of each machine learning model enables the automated remediation platform to monitor the events in real time” contain elements that integrate any alleged abstract idea into a practical application. By using multiple different models, where each model is limited in size and includes curated training data, it is possible for the automated remediation platform to achieve real-time event monitoring. Thus, the features recited in amended claims 1, 11, and 20 clearly address the technical, computer-related challenges associated with "the speed at which the remediation is expected to occur" (see Applicant Remarks, Page 2, dated 08/04/2026). Examiner respectfully disagrees.
Applicant's argument describes an advantage of the claimed arrangement, but does not establish that the claim recites a particular technological solution to the alleged computer problem. In response, Examiner notes that Claims 1, 11 and 20 requires:
"a plurality of machine learning models"
and requires each model to monitor a different combination of subsystems.
But these claims do not require: a particular ML architecture, a particular model-training algorithm, a particular model-partitioning algorithm, a particular model-selection algorithm, a particular model-compression technique, a particular pruning technique, a particular quantization technique, a particular parameter-reduction technique; or a particular process for determining the appropriate subsystem combinations.
Thus, Claims 1, 11 and 20 establishes the existence and intended use of multiple models, but not the technical process by which multiple models produce the asserted computer-performance improvement. The use of multiple models is therefore a functional organization of information-processing resources for performing the abstract enterprise-monitoring activity, rather than a claimed improvement to the underlying ML technology.
Applicant argues "Multiple Models Enable Smaller Models" in Independent Claims 1, 11 and 20 integrate the judicial exception into a practical application or recite additional elements that are significantly more than the judicial exception. Examiner respectfully disagrees. In response, Examiner discloses that this is a particularly important argument because it invokes Enfish and Desjardins-type reasoning. Applicant essentially asserts: Different subsystem combinations → curated data → smaller models → faster processing.
Examiner discloses that these claims do not establish that causal chain with sufficient technical specificity. These claims say: "each machine learning model having a limited size" and: "wherein the limited size ... enables ... monitor[ing] ... in real time." That is essentially a functional relationship between model size and monitoring speed. But these claims do not state: how the models are reduced in size. That distinction is crucial. For example, these claims do not require a particular: model compression technique; pruning process; distillation process; sparse-model construction; reduced-precision representation; feature-reduction technique; parameter-selection process; or hardware implementation.
Accordingly, Examiner points out that these limitations identify a desired model characteristic—limited size—and a desired performance result—real-time monitoring—but does not claim the particular technological mechanism responsible for achieving that result.
Applicant argues that the claim amendments to Independent Claims 1, 11 and 20 " Solves the Computer Problem of Real-Time Monitoring" integrate the judicial exception into a practical application or recite additional elements that are significantly more than the judicial exception. Examiner respectfully disagrees. Applicant relies heavily on paragraph [0045] and states that the amendment solves: "the computer problem how to achieve real-time monitoring."
In response, Examiner notes that the Applicant has identified: real-time event monitoring. But these claims do not necessarily claim the technical solution to achieving real-time monitoring.
Claims 1, 11 and 20 essentially says: use relatively small ML models so that monitoring can occur in real time. That is a result-oriented limitation unless the claim specifies the technological mechanism that produces the result.
Thus, merely recognizing that smaller models can process information more quickly does not itself establish that the claimed arrangement constitutes an improvement to computer functionality.
Applicant's specification states: "the models are relatively small, allowing for quick if not real-time monitoring." Examiner notes that the specification explains the desired result, but these claims still does not specify how the models are technically made small.
Examiner discloses that although paragraph [0045] describes reduced model size and faster monitoring as advantages, the claims do not recite a particular model architecture or technical process that produces the reduced model size. The claims instead require models possessing the desired characteristic of being "limited" in size.
Applicant argues that the claim amendments to Independent Claims 1, 11 and 20 which refer to "Curated Training Data Is Technological" integrate the judicial exception into a practical application or recite additional elements that are significantly more than the judicial exception. Examiner respectfully disagrees. Applicant relies on: "including curated training data specific to the different combination of the plurality of subsystems."
In response, Examiner points out that "Curated training data" identifies the content or selection of information used to train a model, but does not recite a particular technical process for curating, weighting, labeling, preprocessing, or otherwise transforming the training data.
Thes claims do not specify how data is selected; how data is labeled; how data is weighted; how data is filtered; how data is transformed; how data affects model parameters; or
how curation reduces model size. Therefore, the limitation does not necessarily establish an improvement in computer functionality.
Applicant argues that the claim amendments to Independent Claims 1, 11 and 20 which refer to "The Models Are Different for Different Subsystem Combinations" integrate the judicial exception into a practical application or recite additional elements that are significantly more than the judicial exception. Examiner respectfully disagrees.
Applicant contends that this is unlike Recentive Analytics because the claimed models are specifically adapted to different subsystem combinations.
Examiner distinguishes this argument as follows: the claimed specialization identifies what information the models monitor, but does not specify a particular technological improvement in the models themselves.
Claims 1, 11 and 20 establishes:
PNG
media_image1.png
63
277
media_image1.png
Greyscale
but it does not specify the technical structure of Models A, B, and C. Consequently, these limitations are characterized as: dividing information into categories and assigning an analytical model to each category. That remains information processing.
Applicant argues that the claim amendments to Independent Claims 1, 11 and 20 which refer to "The Models Are Smaller Than a One-Size-Fits-All Model” integrate the judicial exception into a practical application or recite additional elements that are significantly more than the judicial exception. Examiner respectfully disagrees.
In response, these claims do not require that the plurality of models actually be smaller than a particular one-size-fits-all model. Instead, it requires: "each machine learning model having a limited size." The claim therefore does not expressly establish a quantitative comparison between the claimed plurality of models; and a hypothetical one-size-fits-all model, nor does it specify: a model-size threshold; a maximum number of parameters; a memory footprint; processing complexity; latency requirement; or computational-resource reduction. Thus, the alleged comparison is primarily found in Applicant's argument and specification, rather than being technically defined in these claims.
Applicant argues that the claim amendments to Independent Claims 1, 11 and 20 which refer to "Limited Model Size Enables Real-Time Monitoring" integrate the judicial exception into a practical application or recite additional elements that are significantly more than the judicial exception. Examiner respectfully disagrees. Examiner notes that these limitations state a functional result without reciting the specific technological mechanism producing the result. The claim says: "wherein the limited size ... enables ... monitor[ing] ... in real time." It does not say why the model is limited in size, nor does it specify how the reduced size changes processor operation. Therefore, these limitations are treated as a desired relationship between model size and processing speed.
Examiner further notes that a claimed relationship between an input characteristic and a desired performance result does not, standing alone, establish a technological improvement when the claim does not recite the particular technical means by which the performance improvement is achieved.
The USPTO's 2025 update specifically recognizes Ex parte Desjardins court case decision on September 26, 2025, and later designated it as precedential on November 4, 2025 as requiring consideration of improvements to learning models and computer functionality.
With respect to the Ex parte Desjardins decision on September 26, 2025, and later designated it as precedential on November 4, 2025 case, the USPTO describes Desjardins as involving ML improvements that: use less storage capacity; reduce system complexity; and allow learning new tasks while preserving prior knowledge. The claims here, by contrast, do not expressly recite a corresponding technical mechanism. They do not claim: a particular way of reducing storage capacity or: a particular learning architecture that preserves prior knowledge or: a particular technical structure that reduces system complexity.
Instead, they claim multiple models having limited size and curated data. Examiner points out that unlike the Ex parte Desjardins case, the claims here do not recite a specific improvement to the internal operation or architecture of an ML system. The asserted improvement is instead the use of appropriately sized and trained models to accomplish the desired enterprise function of real-time failure monitoring. Recentive Analytics, Inc. v. Fox Corp court case issued on April 18, 2025 is highly relevant because it involves machine-learning claims. The Federal Circuit held the claims ineligible because they were directed to using a generic ML technique in a particular environment and lacked an inventive concept.
Examiner argues that like Recentive Analytics, Inc. v. Fox Corp court case issued on April 18, 2025, the present claims employ machine learning to perform a particular function in a particular environment—here, detecting enterprise-system failures and initiating remediation. Applicant's addition of subsystem-specific models; curated data; limited-size models; and real-time monitoring does not necessarily change the fundamental character of the claim because these limitations remain directed to how the abstract enterprise-monitoring objective is accomplished, rather than to a specific improvement in ML technology.
Applicant’s amendments establish that multiple subsystem-specific models may be smaller and may facilitate real-time monitoring, but the claims do not recite the particular technological mechanism by which the models are made smaller or by which their operation improves computer functionality. Rather, the claims specify the desired characteristics and result of the models—limited size, curated training data, and real-time monitoring—and use those models to perform the abstract activity of detecting enterprise failures and initiating remediation. The claimed "curated training data" likewise does not establish a technological improvement because the claims do not recite a particular technique for selecting, processing, weighting, or incorporating the training data that improves operation of the ML system. The "different combination of subsystems" limitation organizes event information and associates particular models with particular categories of events, but does not specify a new ML architecture or computer architecture. The "real-time" limitation states a desired performance result, rather than the technical mechanism responsible for achieving that result. Accordingly, the amended claims remain directed to the abstract enterprise-management concept of automated failure detection and remediation, and the additional computer and ML limitations do not integrate that concept into a practical application under Step 2A Prong Two or provide significantly more under Step 2B.
Independent Claims 1, 11 and 20: With respect to reliance on (e.g., “the at least one of a plurality of subsystems” & “event subsystem” & “device” & “at least one subsystem” & “employee subsystems” & “customer subsystems” & “automated remediation platform” & “a plurality of subsystems” & “computer readable medium” & “enterprise system” & “machine learning models”) as additional elements shown in Independent Claims 1, 11 and 20 when considered individually and as an ordered combination (as a whole) in view of these claim limitations, these additional elements do not provide limitations that are indicative of integration into a practical application under step 2a prong 2 due to: (1) the claims as a whole are limited to a particular field of use or technological environment for generating a trigger event for consumption by the event subsystem and transmitting a remediation event for consumption by the at least one of a plurality of subsystems in a banking or financial institution business enterprise environment (see MPEP § 2106.05 (h)) or (2) recites mere instructions to implement an abstract idea on a computer or using a computer as a tool to “apply” the recited judicial exceptions by providing the results to the user on a computer (see MPEP § 2106.05 (f)).
Moreover, with respect to Independent Claims 1, 11 and 20, certain/particular limitations shown recite (1) mere data gathering (e.g., “receive a request to perform an action, wherein completion of the action involves at least one subsystem of a plurality of subsystems comprising employee subsystems and customer subsystems”) (2) mere data outputting/displaying (e.g., “transmit an event, based on the request, to the at least one subsystem” & “transmit a remediation event for consumption by the at least one subsystem”) in which each of these claim limitations reflects mere insignificant extra-solution activities (see MPEP § 2106.05 (g)). In addition, these limitations fail to provide an improvement to the functioning of a computer or to any other technology or technical field, fail to apply the exception with a particular machine, fail to apply the judicial exception to effect a particular treatment or prophylaxis for a disease or medical condition, fail to effect a transformation of a particular article to a different state or thing, and fail to apply/use the abstract idea in a meaningful way beyond generally linking the use of the judicial exception to a particular technological environment. Accordingly, because the Step 2A Prong One and Prong Two analysis resulted in the conclusion that the claims are directed to an abstract idea, additional analysis under Step 2B of the eligibility inquiry must be conducted in order to determine whether any claim element or combination of elements amount to significantly more than the judicial exception.
Argument #2:
(B). Applicant argues that Dependent Claims 23-24 which depends from Independent Claims 11 and 20 recite additional elements that integrate the judicial exception into a practical application under revised step 2a prong two of the 35 U.S.C. § 101 analysis (see Applicant Remarks, Page 3, dated 08/04/2026). Examiner respectfully disagrees.
Examiner responds by stating that Dependent Claims 23-24 that the action of training and updating a model to recognize patterns of failure falls under the abstract idea groupings of a "mental process" or "mathematical concept/data analysis. Under Step 2A, Prong 2, the claim fails to integrate the exception into a practical application. Updating the platform: Merely using a computer system or platform as a tool to store and run the model is insignificant extra-solution activity. Training a new ML model: Generically "training" or "re-training" an ML model on a new data set (detecting a different failure) is incident to the nature of machine learning and does not improve the internal operation or functioning of the computer or the ML algorithm itself. Detecting a different type of failure: Limiting the abstract concept of pattern recognition to a specific field of use (a new type of system failure) does not steer the claim away from being "directed to" an abstract idea. Utilizing training steps, iterative data parsing, and feedback loops do not amount to an inventive concept because they represent activities in the art. The combination of receiving failure data, processing it through a standard ML algorithm, and outputting an updated detection profile lacks a specific, unconventional technical solution to a technical problem in computer science. It merely automates a diagnostic task that was previously manual or rule-based, rendering the claim ineligible. There is no "inventive concept" because the steps are structured in a manner to automate an analytical process. These limitations fail Step 2B. It does not provide "significantly more" than the underlying abstract idea, rendering Claims 23-24 patent-ineligible under 35 U.S.C. § 101.
Therefore, the ordered combination of elements in the Dependent Claims (including the limitations inherited from the parent claim(s)) add nothing that is not already present as when the elements are taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Accordingly, the subject matter encompassed by the dependent claims fails to amount to a practical application or significantly more than the abstract idea itself. Therefore, under Step 2B, Claims 1, 4-11 and 14-24 do not include additional elements that are sufficient to amount to significantly more than the recited judicial exceptions. Thus, Claims 1, 4-11 and 14-24 are maintained as being patent ineligible with respect to the 35 U.S.C. § 101 analysis.
Claim Rejections - 35 USC § 101
6. 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.
7. Claims 1, 4-11 and 14-24 are rejected under 35 U.S.C. § 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1: Claims 1, 4-11 and 14-24 are focused to a statutory category namely, a “device” or a “system” (Claims 1, 4-10 and 21-22), a “method” or a “process” (Claims 11, 14-19 and 22-23) and a “non-transitory computer-readable medium” or an “article of manufacture” (Claims 20 and 24).
Step 2A Prong One: Independent Claims 1, 11 and 20 recites limitations that set forth the abstract idea(s), namely (see in bold except where strikethrough):
“” (see Independent Claim 1);
“” (see Independent Claim 1);
“for at least in part automating remediation in an enterprise
“receive a request to perform an action, wherein completion of the action involves comprising employee and customer ” (see Independent Claims 1, 11 and 20);
“transmit an event, based on the request, via an event , wherein the event comprises a dedicated notification event manager for managing notification events and a dedicated action event manager for managing action events” (see Independent Claims 1, 11 and 20);
“at an automated remediation ” (see Independent Claims 1, 11 and 20);
“store a plurality of models trained to detect failures, each model having been trained to monitor events that involve a different combination and each model having a limited size and including curated training data specific to the different combination ” (see Independent Claims 1, 11 and 20);
“using the plurality of models, monitor events, in real time, sent from to the event to detect a failure, wherein the limited size of each model enables to monitor the events in real time” (see Independent Claims 1, 11 and 20);
“in response to detecting the failure, generate a trigger event for consumption by the event ” (see Independent Claims 1, 11 and 20);
“in response to receiving the trigger event at the event , transmit a remediation event for consumption , wherein the remediation event is one of (i) a notification event handled by the dedicated notification event manager; and (ii) an action event handled by the dedicated action event manager” (see Independent Claims 1, 11 and 20).
Here, for Independent Claims 1, 11 and 20, the abstract ideas recite the mental and organizational concepts of collecting data, analyzing information using statistical/machine learning models to detect events or failures, and generating notification or response actions based on that analysis. More specifically, the core concept of the abstract idea is the fundamental mental process of monitoring system activity, running data-based classification or detection rules (machine learning models) on grouped subsets of data, and triggering conditional messages or corrective workflows. This is the digital equivalent of a human supervisor or dispatcher watching multiple department logs (employee/customer streams), spotting an error, and deciding whether to send an alert or an instruction to fix it. In summary, these claim limitations describe a system with a processor and memory that receives a request, sends events through a notification and action manager, uses small machine learning models to monitor subsystem failures in real time, and triggers a remediation event (notification or action). Under 35 U.S.C. 101, it is directed to the abstract idea of monitoring system events and managing automated alerts or responses to failures.
For example; in Independent Claims 1, 11 and 20 regarding the first claim step of “Receiving a Request” via “...receive a request to perform an action, wherein completion of the action involves at least one subsystem of a plurality of subsystems comprising employee subsystems and customer subsystems..." recites “data collection and receiving input data” which is an abstract, generic concept of gathering information. This falls under Certain Method of Organizing Human Activity (data gathering / business interaction).
The second claim limitation step of “Transmitting an Event via an Event Subsystem” via "...transmit an event, based on the request, to the at least one subsystem via an event subsystem, wherein the event subsystem comprises a dedicated notification event manager for managing notification events and a dedicated action event manager for managing action events..." recites categorizing, routing, and transmitting messages/events using specialized managers describes organizational communication structures and administrative workflow rules. This falls under Certain Method of Organizing Human Activity (data routing / communication management).
Next, the third claim limitation step of “Storing Machine Learning Models with Curated Data” via "...at an automated remediation platform ; store a plurality of machine learning models trained to detect failures, each machine learning model having been trained to monitor events that involve a different combination of the plurality of subsystems, and each machine learning model having a limited size and including curated training data specific to the different combination of the plurality of subsystems..." recites mathematical calculations, data analysis, and training statistical models on selected data subsets which fall squarely under mathematical concepts and abstract data analysis. This falls under Mental Process (mental concepts / generalized data analysis techniques) or Math Concepts (statistical/mathematical modeling).
The fourth claim limitation step of “Real-Time Monitoring using ML Models” of "...using the plurality of machine learning models, monitor events, in real time, sent from the at least one subsystem to the event subsystem to detect a failure, wherein the limited size of each machine learning model enables the automated remediation platform to monitor the events in real time..." recites the concept of "monitoring in real time" via data evaluation which is a mental/analytical process expedited by processing speed. This falls under Mental Process (evaluating information, diagnosing conditions). The fifth step of “Generating a Trigger Event” via "...in response to detecting the failure, generate a trigger event for consumption by the event subsystem..” recites a conditional generation of a signal upon satisfying a logical threshold is a fundamental cause-and-effect rule. This falls under Mental Process / Certain Method of Organizing Human Activity (decision-making based on analyzed data). Furthermore, the last step of “Transmitting a Remediation Event (Notification or Action)” via "...in response to receiving the trigger event at the event subsystem, transmit a remediation event for consumption by the at least one subsystem, wherein the remediation event is one of (i) a notification event handled by the dedicated notification event manager; and (ii) an action event handled by the dedicated action event manager..." recites delivering an output notification or action directive based on pre-determined rules which an administrative communication step. This falls under Certain Method of Organizing Human Activity (administering a communication or workflow rule).
A human operator or system administrator can mentally or manually observe incoming reports from different departments (subsystems), note when a failure occurs based on past experiences or training guidelines (analogous to the ML models), and decide to send an alert or to fix instruction. Managing communication flow, routing notifications, and handling business workflow actions are commercial and organizational practices. The recitation of a "processor," "memory," and "event subsystem" merely applies to these mental and organizational concepts using computer hardware and networking components.
Therefore, these abstract idea limitations (as identified above in bold), under their broadest reasonable interpretation of the claims as a whole, cover performance of their limitations as “Certain Methods of Organizing Human Activities” which pertains to (1) managing personal behavior or relationships or interactions between people (including teachings or following rules or instructions) or (2) fundamental economic principles or practices (including mitigating risk) or (3) commercial interactions (including business relations).
Additionally, or alternatively, these abstract idea limitations (as identified above in bold), under their broadest reasonable interpretation of the claims as a whole, cover performance of their limitations as “Mental Processes” which pertains to (4) concepts performed in the human mind (including observations or evaluations or judgments) or (5) using pen and paper as a physical aid, which in order to help perform these mental steps does not negate the mental nature of these limitations. The use of "physical aids" in implementing the abstract mental process, does not preclude the claim from reciting an abstract idea. See MPEP § 2106.04(a) III C.
That is, other than reciting (e.g., “the at least one of a plurality of subsystems” & “event subsystem” & “device” & “at least one subsystem” & “employee subsystems” & “customer subsystems” & “automated remediation platform” & “a plurality of subsystems” & “computer readable medium” & “enterprise system” & “a processor” & “a memory”, etc…), nothing in the claim elements precludes the steps from being performed as “Certain Methods of Organizing Human Activities” which pertains to (1) managing personal behavior or relationships or interactions between people (including teachings or following rules or instructions) or (2) fundamental economic principles or practices (including mitigating risk) or (3) commercial interactions (including business relations) and additionally or alternatively as “Mental Processes” which pertains to (4) concepts performed in the human mind (including observations or evaluations or judgments) or (5) using pen and paper as a physical aid.
Therefore, at step 2a prong 1, Yes, Claims 1, 4-11 and 14-24 recites an abstract idea. We proceed onto analyzing the claims at step 2a prong 2.
Step 2A Prong Two: With respect to Step 2A Prong Two of the eligibility inquiry (as explained in MPEP § 2106.04(d)), the judicial exception is not integrated into a practical application. Independent Claim 1 recites additional elements directed to: (e.g., “a processor” & “a memory”). These additional elements have been considered individually and in combination, but fail to integrate the abstract idea into a practical application because they amount to using computing elements or instructions (software) to perform the abstract idea, similar to adding the words “apply it” (or an equivalent), which merely serves to link the use of the judicial exception to a particular technological environment. See MPEP § 2106.05(f) and MPEP § 2106.05(h).
Independent Claims 1, 11 and 20: With respect to reliance on (e.g., “the at least one of a plurality of subsystems” & “event subsystem” & “device” & “at least one subsystem” & “employee subsystems” & “customer subsystems” & “automated remediation platform” & “a plurality of subsystems” & “computer readable medium” & “enterprise system” & “machine learning models”) as additional elements shown in Independent Claims 1, 11 and 20 when considered individually and as an ordered combination (as a whole) in view of these claim limitations, these additional elements do not provide limitations that are indicative of integration into a practical application under step 2a prong 2 due to: (1) the claims as a whole are limited to a particular field of use or technological environment for generating a trigger event for consumption by the event subsystem and transmitting a remediation event for consumption by the at least one of a plurality of subsystems in a banking or financial institution business enterprise environment (see MPEP § 2106.05 (h)) or (2) recites mere instructions to implement an abstract idea on a computer or using a computer as a tool to “apply” the recited judicial exceptions by providing the results to the user on a computer (see MPEP § 2106.05 (f)).
Moreover, with respect to Independent Claims 1, 11 and 20, certain/particular limitations shown recite (1) mere data gathering (e.g., “receive a request to perform an action, wherein completion of the action involves at least one subsystem of a plurality of subsystems comprising employee subsystems and customer subsystems”) (2) mere data outputting/displaying (e.g., “transmit an event, based on the request, to the at least one subsystem” & “transmit a remediation event for consumption by the at least one subsystem”) in which each of these claim limitations reflects mere insignificant extra-solution activities (see MPEP § 2106.05 (g)). In addition, these limitations fail to provide an improvement to the functioning of a computer or to any other technology or technical field, fail to apply the exception with a particular machine, fail to apply the judicial exception to effect a particular treatment or prophylaxis for a disease or medical condition, fail to effect a transformation of a particular article to a different state or thing, and fail to apply/use the abstract idea in a meaningful way beyond generally linking the use of the judicial exception to a particular technological environment. Accordingly, because the Step 2A Prong One and Prong Two analysis resulted in the conclusion that the claims are directed to an abstract idea, additional analysis under Step 2B of the eligibility inquiry must be conducted in order to determine whether any claim element or combination of elements amount to significantly more than the judicial exception. Therefore, at step 2a prong 2, Claims 1, 4-11 and 14-24 are directed to the abstract idea and do not recite additional elements that integrate into a practical application.
Step 2B: (As explained in MPEP § 2106.05), it has been determined that the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. Independent Claim 1 recites additional elements directed to: (e.g., “a processor” & “a memory”). These elements have been considered individually and in combination, but fail to add significantly more to the claims because they amount to using computing elements or instructions (software) to perform the abstract idea, similar to adding the words “apply it” (or an equivalent), which merely serves to link the use of the judicial exception to a particular technological environment (computing environment) and does not amount to significantly more than the abstract idea itself. Notably, Applicant’s Specification suggests that the claimed invention relies on nothing more than a computer executing the instructions to implement the invention (see at least Applicant’s Specification ¶ [0107]: “The user device 12 includes a display module 164 for rendering GUIs and other visual outputs on a display device such as a display screen, and an input module 1866 for processing user or other inputs received at the user device 12, .g., via a touchscreen, input button, transceiver, microphone, keyboard.”)
Independent Claims 1, 11 and 20: With respect to reliance on (e.g., “the at least one of a plurality of subsystems” & “event subsystem” & “device” & “at least one subsystem” & “employee subsystems” & “customer subsystems” & “automated remediation platform” & “a plurality of subsystems” & “computer readable medium” & “enterprise system” & “machine learning models”) as additional elements shown in Independent Claims 1, 11 and 20 when considered individually and as an ordered combination (as a whole) in view of these claim limitations, these additional elements do not recite additional elements that amount to significantly more than the recited judicial exceptions under step 2B due to: (1) the claims as a whole are limited to a particular field of use or technological environment for generating a trigger event for consumption by the event subsystem and transmitting a remediation event for consumption by the at least one of a plurality of subsystems in a banking or financial institution business enterprise environment (see MPEP § 2106.05 (h)) or (2) recites mere instructions to implement an abstract idea on a computer or using a computer as a tool to “apply” the recited judicial exceptions by providing the results to the user on a computer (see MPEP § 2106.05 (f)).
For Independent Claims 1, 11 and 20, the step of “receiving a request to perform an action, wherein completion of the action involves at least one subsystem of a plurality of subsystems comprising employee subsystems and customer subsystems” recites data gathering and receiving user/system requests involving administrative categories (employee vs. customer). This falls squarely under the abstract idea categories of commercial/business interactions or data sorting/receiving. It is recited at a high, functional level without defining a specific, unconventional technological protocol or improvement in data transfer. Receiving requests from disparate user interfaces or functional silos is a conventional enterprise task. It lacks any inventive concept beyond standard data intake. The step of “transmit an event, based on the request, to the at least one subsystem via an event subsystem, wherein the event subsystem comprises a dedicated notification event manager for managing notification events and a dedicated action event manager for managing action events” describes the generic organization of tasks—separating notification management from action management. Categorizing and routing messages based on type is an abstract concept of data classification and administrative organization. The use of "dedicated managers" is purely functional language rather than a specific hardware architecture improvement. Dividing event-driven architectures into discrete managers or handlers for notifications versus actions is a conventional software design pattern (e.g., publisher-subscriber models, separation of concerns). It adds no inventive concept.
The step of “an automated remediation platform; store a plurality of machine learning models trained to detect failures, each machine learning model having been trained to monitor events that involve a different combination of the plurality of subsystems, and each machine learning model having a limited size and including curated training data specific to the different combination of the plurality of subsystems” is analyzed as storing models and training them on curated data sets to recognize patterns or failures represents an abstract idea of mental processes, data analysis, or mathematical calculations/machine learning. While models are specialized to specific subsystem combinations, this represents application of mathematical concepts to a business/system environment rather than an improvement to computer data processing capabilities itself. Training machine learning models on specific data subsets (curated training data) is performed in data science. Limiting model size for performance is an ordinary engineering choice, not an inventive concept. The step of “using the plurality of machine learning models, monitor events, in real time, sent from the at least one subsystem to the event subsystem to detect a failure, wherein the limited size of each machine learning model enables the automated remediation platform to monitor the events in real time” recites "Monitoring in real time" using statistical or machine learning classification is an abstract process of observation and diagnosis. Attributing real-time capability to the "limited size" of the model is a restatement of a natural result (smaller models run faster) rather than a technical solution to a technical problem rooted in computer architecture. Applying generic predictive models to incoming data streams to flag anomalies or failures is standard practice. It provides no unconventional technological transformation. The step of “in response to detecting the failure, generate a trigger event for consumption by the event subsystem” recites a conditional generation of an alert or trigger upon detecting a threshold/failure is a fundamental, abstract thought process (if condition X, generate alert Y). It does not integrate the data into a concrete, non-abstract technical output. Generating triggers or flags based on algorithmic outcomes is utilized in data processing. The step of in response to receiving the trigger event at the event subsystem, transmit a remediation event for consumption by the at least one subsystem, wherein the remediation event is one of (i) a notification event handled by the dedicated notification event manager; and (ii) an action event handled by the dedicated action event manager” is noted as a concluding step mapping back to routing messages (notifications vs. actions) in response to a trigger. It represents generic automated communication and data dispatching—an abstract idea without an inventive transformation. Routing a response event to an appropriate handler based on the event type is a routine automation that lacks an inventive concept.
Moreover, with respect to Independent Claims 1, 11 and 20, certain/particular limitations shown recite (1) mere data gathering (e.g., “receive a request to perform an action, wherein completion of the action involves at least one subsystem of a plurality of subsystems comprising employee subsystems and customer subsystems”) (2) mere data outputting/displaying (e.g., “transmit an event, based on the request, to the at least one subsystem” & “transmit a remediation event for consumption by the at least one subsystem”) in which each of these claim limitations reflects mere insignificant extra-solution activities (see MPEP § 2106.05 (g)). Furthermore, these certain/particular limitations claim limitations as demonstrated above for Independent Claims 1, 11 and 20 reflect Well-Understood, Routine and Conventional Activities (WURC) under MPEP § 2106.05 (d) ii: See Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec,838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359,1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network). Moreover, with respect to Independent Claims 1, 11 and 20, certain/particular limitations shown recite storing data such as (e.g., “store a plurality of machine learning models trained to detect failures, each machine learning model having been trained to monitor events that involve a different combination of the plurality of subsystems, and each machine learning model having a limited size and including curated training data specific to the different combination of the plurality of subsystems”) reflect Well-Understood, Routine and Conventional Activities (WURC) under MPEP § 2106.05 (d) ii: See Storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc.,793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363, 115USPQ2d at 1092-93.
In addition, when taken as an ordered combination, the ordered combination adds nothing that is not already present as when the elements are taken individually. There is no indication that the combination of elements integrates the abstract idea into a practical application. Therefore, when viewed as a whole, these additional claim elements do not provide meaningful limitations to transform the abstract idea into a practical application of the abstract idea or that, as an ordered combination, amount to significantly more than the abstract idea itself.
Dependent Claims 4-10, 14-19 and 21-24 recite additional elements directed to: (e.g. “enterprise account subsystem” (Dependent Claims 4 and 14) & “wealth management” (Dependent Claims 4 and 14) & “payments” (Dependent Claims 4 and 14) & “an employee subsystem” (Dependent Claims 5 and 15) & “automated event manager” (Dependent Claims 8 and 18) & “notification manager” (Dependent Claims 8 and 18) & “new machine learning model” (Dependent Claim 10) & “the plurality of machine learning models” (Dependent Claims 21-22) & “automated remediation platform” (Dependent Claims 23-24) & “new machine learning model” (Dependent Claims 23-24)), which in conjunction with the limitations recite the same abstract idea(s) as shown in Independent Claims 1, 11 and 20 along with further steps/details that reflect “Certain Methods of Organizing Human Activities” which pertains to (1) managing personal behavior or relationships or interactions between people (including teachings or following rules or instructions) or (2) fundamental economic principles or practices (including mitigating risk) or (3) commercial interactions (including business relations) and additionally or alternatively as “Mental Processes” which pertains to (4) concepts performed in the human mind (including observations or evaluations or judgments) or (5) using pen and paper as a physical aid.
Dependent Claims 6-7, 9, 6-17 and 19 further narrow the abstract ideas, and are therefore still ineligible for the reasons previously provided in Steps 2A Prong 2 and Step 2B for Independent Claims 1, 11 and 20. Dependent Claims 4-5, 8, 10, 14-15, 18 and 21-24: With respect to reliance on (e.g. “enterprise account subsystem” (Dependent Claims 4 and 14) & “wealth management” (Dependent Claims 4 and 14) & “payments” (Dependent Claims 4 and 14) & “an employee subsystem” (Dependent Claims 5 and 15) & “automated event manager” (Dependent Claims 8 and 18 ) & “notification manager” (Dependent Claims 8 and 18) & “new machine learning model” (Dependent Claim 10) & “the plurality of machine learning models” (Dependent Claims 21-22) & “automated remediation platform” (Dependent Claims 23-24) & “new machine learning model” (Dependent Claims 23-24)) as additional elements shown in Dependent Claims 4-5, 8, 10, 14-15, 18 and 21-24 when considered both individually and as an ordered combination (as a whole) with these recited claim limitations, these additional elements do not provide limitations that are indicative of integration into a practical application under step 2a prong 2 and also do not amount to significantly more than the judicial exceptions under step 2B due to the following: (1) the claims as a whole are limited to a particular field of use or technological environment for generating a trigger event for consumption by the event subsystem and transmitting a remediation event for consumption by the at least one of a plurality of subsystems in a banking or financial institution business enterprise environment (see MPEP § 2106.05 (h)) or (2) recites mere instructions to implement an abstract idea on a computer or using a computer as a tool to “apply” the recited judicial exceptions by providing the results to the user on a computer (see MPEP § 2106.05 (f)).
Dependent Claims 21-22: These limitations describe a purely functional result (achieving a trained model) rather than a specific structural or algorithmic mechanism. "incident to the very nature of machine learning" and does not represent an eligible technological advancement on its own. Under the USPTO's expanded guidelines, a training step only establishes an improvement if it alters the internal operation of the model to solve an inherent computing deficit (e.g., preventing "catastrophic forgetting", decreasing parameter storage size, or optimizing mathematical convergence mechanics). Simply retraining a model using protocols fails to improve the underlying computer or model functionality. This sub-step fails Prong 2 because it applies training techniques at a high level.
Feeding updated datasets into an existing model layout is a data-gathering and data-feeding activity. It changes the data inputs, but it does not change the core architecture of the software or computing hardware. The court ruled that "applying established methods of machine learning to a new data environment... is patent ineligible under § 101." Swapping out static data for "updated" or "real-time" data is simply updating a data environment. This sub-step fails Prong 2 because it limits the abstract process to a particular field/data type, which does not constitute a technical integration. The "automated remediation platform" serves merely as a generic, high-level computer environment where the abstract mathematical step takes place. The step does not describe how the platform handles hardware resources more effectively. Limiting an abstract idea to a particular technology platform or field of use (remediation) does not transition an abstract concept into an eligible practical application. Automation of a task via generic computer components does not create eligibility, even if it introduces vast gains in execution speed or computational efficiency. This sub-step fails Prong 2 because the platform acts as a generic computer environment implementing the exception. These limitations fails Step 2A, Prong 2. The combination of steps does not optimize computer operation or resolve a technical bottleneck. It merely applies machine learning workflows to generic data.
These elements of "updating the automated remediation platform" are an instruction to execute software code on generic hardware. It is an element that adds nothing inventive. "retraining... machine learning models" is noted as training loops (e.g., backpropagation, gradient descent) which are baseline protocols across the software engineering field. The claim does not introduce a novel algorithmic structure. "Using updated training data": Data collection, ingest optimization, and memory loading are baseline, standard functions of any data processing apparatus. When viewed as an ordered combination, the sequence of events describes a standard data cycle: Collect new data → Feed it into a generic model to retrain weights → Update the host platform software. There is no "inventive concept" because the steps are structured in a manner to automate an analytical process. These limitations fail Step 2B. It does not provide "significantly more" than the underlying abstract idea, rendering Claims 21-22 patent-ineligible under 35 U.S.C. § 101.
Dependent Claims 23-24: The action of training and updating a model to recognize patterns of failure falls under the abstract idea groupings of a "mental process" or "mathematical concept/data analysis. Under Step 2A, Prong 2, the claim fails to integrate the exception into a practical application. Updating the platform: Merely using a computer system or platform as a tool to store and run the model is insignificant extra-solution activity. Training a new ML model: Generically "training" or "re-training" an ML model on a new data set (detecting a different failure) is incident to the nature of machine learning and does not improve the internal operation or functioning of the computer or the ML algorithm itself. Detecting a different type of failure: Limiting the abstract concept of pattern recognition to a specific field of use (a new type of system failure) does not steer the claim away from being "directed to" an abstract idea. Utilizing training steps, iterative data parsing, and feedback loops do not amount to an inventive concept because they represent activities in the art. The combination of receiving failure data, processing it through a standard ML algorithm, and outputting an updated detection profile lacks a specific, unconventional technical solution to a technical problem in computer science. It merely automates a diagnostic task that was previously manual or rule-based, rendering the claim ineligible. There is no "inventive concept" because the steps are structured in a manner to automate an analytical process. These limitations fail Step 2B. It does not provide "significantly more" than the underlying abstract idea, rendering Claims 23-24 patent-ineligible under 35 U.S.C. § 101.
The additional element of “machine learning models” for these claims does not amount to significantly more than the judicial exception under step 2B due to being expressly recognized as known in the art. See for example., US PG Pub (US 2023/0126193 A1) hereinafter Nowak, et. al. Nowak at ¶ [0053]: This user confirmation and/or user override of remediation action assignment may be feedback data to the machine learning model data store 311 and/or new incident data 301. Data maintained in the new incident data 301 and utilized by the machine learning model 331 described herein may be updated to account for the confirmation data 361. Such an update may include creating, in the database maintaining new incident data, a new database entry comprising the assigned remediation actions and the one or more other assets of the entity that have not been subjected to the specific incident. Nowak at ¶ [0007]: Generally, enable predicting similarities, in incident data, of incidents that, for a first asset of the entity, were reviewed and had remediation actions assigned to them. The remediation actions may be assigned to one or more second assets of the entity. Nowak at ¶ [0028]: Conventional systems are susceptible to failure or repetition of occurrence of a previous incident—for example, an incident that may occur similarly for another entity resource under a similar situation as an incident that had remediation actions assigned to mitigate reoccurrence of that incident may lead to wasted time and resources to address the occurrence of an incident. As such, these conventional techniques leave entities exposed to the possibility of a constant reoccurrence of the incident on the operation of the entity. By providing improved assignment techniques—for example, based on predicting the likely remediation actions to assign to mitigate occurrence of an incident—a proper remediation action assignment can be more accurately determined.
The ordered combination of elements in the Dependent Claims (including the limitations inherited from the parent claim(s)) add nothing that is not already present as when the elements are taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Accordingly, the subject matter encompassed by the dependent claims fails to amount to a practical application or significantly more than the abstract idea itself. Therefore, under Step 2B, Claims 1, 4-11 and 14-24 do not include additional elements that are sufficient to amount to significantly more than the recited judicial exceptions. Thus, Claims 1, 4-11 and 14-24 are ineligible with respect to the 35 U.S.C. § 101 analysis.
Conclusion
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 DERICK HOLZMACHER whose telephone number is (571) 270-7853. The examiner can normally be reached on Monday-Friday 9:00 AM – 6:30 PM EST.
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, Brian Epstein can be reached on 571-270-5389. The fax phone number for the organization where this application or proceeding is assigned is 571-270-8853.
Information regarding the status of an application may be obtained from Patent Center. Status information for published applications may be obtained from Patent Center. Status information for unpublished applications is available through Patent Center for authorized users only. Should you have questions about access to Patent Center, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
/DERICK J HOLZMACHER/Patent Examiner, Art Unit 3625A
/BRIAN M EPSTEIN/Supervisory Patent Examiner, Art Unit 3625
Prosecution Insights