Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
DETAILED ACTION
Status of the Application
This action is in response to the Amendment filed on 7/29/2026, and is a Final Office Action. Claims 1-25 are pending in the application.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-25 are rejected under 35 U.S.C. § 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Claim 1 is directed towards a method, thus meeting the Step 1 eligibility criterion. Claim 1 does recite the abstract concept of a commercial interaction/fundamental economic practice, which has been identified as an abstract idea by the MPEP. The relevant claimed limitations include: obtaining, prior to an advertising auction, network performance data that , at least in part, characterizes a first performance of one or more of a plurality of available neural-network systems / obtaining, prior to the advertising auction, text engine performance data that, at least in part, characterizes a second performance of a plurality of text engines / obtaining, prior to the advertising auction, imagery engine performance data that, at least in part, characterize a third performance of a plurality of imagery engines / obtaining, during the advertising auction, context data that, at least in part, characterizes an active user session within a digital media environment / generating a plurality of prompt datasets based, at least in part, on the context data, the plurality of prompt datasets comprising at least a first prompt dataset to generate text and a second prompt dataset to generate imagery/ selecting a text engine from the plurality of text engines and selecting a first neural -network computing system from the plurality of available neural-network computing systems based on the network performance data and the text engine performance data / selecting an imagery engine from the plurality of imagery engines and selecting a second neural-network computing system from the plurality of available neural-network computing systems based on the network performance data and the imagery engine performance data /
combining the at least one text object and the at least one visual-collateral object into an ad- copy object formatted for insertion into an advertising slot of a presentation in the active user session, wherein the at least one text object and/or the at least one visual-collateral object is at least partially created after obtaining the context data and before the presentation is presented in the active user session / transmitting a bid in a bid message / in a third operation, in response to acceptance of the transmitted bid, transmitting the ad-copy object for presentation to the user/ determining whether the ad-copy object provides text and imagery detail within a quality score range established by the first prompt dataset, the second prompt dataset, the context data, and a predictive model of what would satisfy a request derived from the first set of training data, wherein the quality score encompasses greater amounts of detail data than explicitly specified in the first prompt dataset and the second prompt dataset. This judicial exception is not integrated into a practical application. Claim 1 includes the additional elements of neural-network computing systems, text engines executable on the neural -network computing systems / imagery engines executable on the neural-network computing systems /ad auction server, which represent generic computing elements. The additional elements of using neural networks to generate data/using prompts /training neural networks/reconfiguring the neural networks (the second prompt dataset is different than the first prompt dataset/ the first neural -network computing system is trained on a first set of training data / providing the first prompt dataset to the text engine executing on the first neural-network computing system, while the text engine is configured to generate text from a text engine prompt / receiving at least one text object, usable as ad-copy text, from the text engine / providing the second prompt dataset to the imagery engine executing on the second neural-network computing system and while the imagery engine is configured to generate visual-collateral objects from an imagery engine prompt / receiving at least one visual-collateral object, usable as ad-copy imagery, from the imagery engine / the second neural-network computing system is trained on a second set of training data / in response to the ad-copy object providing the text and imagery detail outside the quality score range: (a) providing a first reconfiguration signal to the first neural-network computing system and a second set of reference data, wherein the first reconfiguration signal causes the first neural-network computing system to reconfigure in response to the second set of reference data; (b) providing a second reconfiguration signal to the second neural-network computing system and the second set of reference data, wherein the second reconfiguration signal causes the second neural-network computing system to reconfigure in response to the second set of reference data; (c) providing the first reconfiguration signal and the second reconfiguration signal to the first neural-network computing system and the second neural-network computing system in order to reconfigure the first neural-network computing system and reconfigure the second neural-network computing system for use in generating subsequent ad-copy objects with text and imagery detail within the quality score range) do no more than apply or link the use of the recited judicial exception to a particular technological environment/field of use. The additional elements do not , alone or in combination. improve the functioning of the computing device or another technology/technical field, or apply or use the judicial exception in some other meaningful way beyond generally linking its use to a particular technological environment. The claim is directed to an abstract idea. Claim 1 does not include additional elements that are sufficient to amount to significantly more than the judicial exception, because as noted above, the claimed computing elements represent generic computing elements; they are recited at a high level of generality. The additional elements of using neural networks to generate data /using prompts /training neural networks/reconfiguring the neural networks do no more than apply or link the use of the recited judicial exception to a particular technological environment/field of use. The additional elements do not , alone or in combination improve the functioning of the computing device or another technology/technical field, or apply or use the judicial exception in some other meaningful way beyond generally linking its use to a particular technological environment. Therefore, Claim 1 does not amount to significantly more than the abstract idea itself. The claim is not patent eligible.
Claim 13 is directed towards a method, thus meeting the Step 1 eligibility criterion. Claim 13 performs similar claimed limitations to those of claim 1 using only generic components of a networked computer system; the claim recites the same abstract idea as claim 1. Therefore, claim 13 is directed to an abstract idea without significantly more for the reasons given in the discussion of claim 1.
Claim 19 is directed towards a method, thus meeting the Step 1 eligibility criterion.
Claim 19 performs similar claimed limitations to those of claim 1 using only generic components of a networked computer system; the claim recites the same abstract idea as claim 1. Therefore, claim 19 is directed to an abstract idea without significantly more for the reasons given in the discussion of claim 1.
Claim 23 is directed towards a method, thus meeting the Step 1 eligibility criterion. Claim 23 does recite the abstract concept of a commercial interaction/fundamental economic practice, which has been identified as an abstract idea by the MPEP. The relevant claimed limitations include:
obtaining, in response to an advertising request relating to a digital media presentation to a user in a digital media environment to a user, context data characterizing the digital media environment and/or the user; receiving, as part of digital media presentation to be presented to the user, an auction window timing constraint associated with an advertising auction associated with the advertising request, wherein the auction window timing constraint constrains a time period in which to execute an auction for an advertising slot in the digital media presentation, provide a first media object for use in the advertising slot, and present the digital media presentation to the user with the first media object included, wherein the first media object includes at least one text component and at least one visual component; within the time period, providing the first media object to an ad evaluation engine, wherein the ad evaluation engine is configured to execute a plurality of evaluation operations on the first media object to determine whether the first media object meets at least one advertising quality threshold to generate one or more evaluation outputs; determining, based on the one or more evaluation outputs of the plurality of evaluation operations, whether first media object satisfies one or more acceptance criteria; in response to determining that the first media object does not satisfy the one or more acceptance criteria, iteratively modifying at least one of the context data, one or more prompt datasets derived from the context data, or one or more generation parameters to modify the first media object; in response to determining that the first media object does satisfy the one or more acceptance criteria, transmitting a bid associated with the first media object, wherein transmitting occurs within the time period; and upon acceptance of the bid, transmitting the first media object for use in the digital media presentation to the user. This judicial exception is not integrated into a practical application. Claim 23 includes the additional elements of generating data using a generative machine learning model / modifying the machine learning models based on criteria (generating, within the time period and using one or more generative machine-learning models executed on one or more computing systems, the media object / in response to determining that the first media object does not satisfy the one or more acceptance criteria, modifying, within the time period, the one or more generative machine-learning models), a computing system, an ad evaluation engine, an ad auction server. The server/system represent generic computing elements. The additional elements of using a generative machine learning model to generate data/modifying the machine learning models based on criteria do no more than apply or link the use of the recited judicial exception to a particular technological environment/field of use. The additional elements do not , alone or in combination. improve the functioning of the computing device or another technology/technical field, or apply or use the judicial exception in some other meaningful way beyond generally linking its use to a particular technological environment. The claim is directed to an abstract idea. Claim 23 does not include additional elements that are sufficient to amount to significantly more than the judicial exception, because as noted above, the claimed computing elements represent generic computing elements; they are recited at a high level of generality. Using a generative machine learning model to generate data/ modifying the machine learning models based on criteria do no more than apply or link the use of the recited judicial exception to a particular technological environment/field of use. The additional elements do not , alone or in combination improve the functioning of the computing device or another technology/technical field, or apply or use the judicial exception in some other meaningful way beyond generally linking its use to a particular technological environment. Therefore, Claim 23 does not amount to significantly more than the abstract idea itself. The claim is not patent eligible.
Remaining dependent claims 2-12, 14-18, 20, 21, 22, 24, 25 further recite and narrow the abstract ideas of the independent claims themselves. The claims further recite the additional elements of a real-time programmatic platform, using a generative machine-learning model to generate data, and a hardware system/text engine/imagery engine, determining data based on an embedding-based language model. Processing data within a real-time platform does no more than apply or link the use of the recited judicial exception to a particular technological environment. Using a generative machine-learning model to generate data does no more than apply or link the use of the recited judicial exception to a particular technological environment. The hardware system/claimed engines represent generic computing elements that are recited at a high level of generality. Determining data based on an embedding-based language model does no more than apply or link the use of the recited judicial exception to a particular technological environment. The additional elements do not, alone or in combination with the other additional elements, improve the functioning of the computing device or another technology/technical field, or apply or use the judicial exception in some other meaningful way beyond generally linking its use to a particular technological environment. Therefore, the claims do not amount to significantly more than the abstract idea itself. The claims are not patent eligible.
The prior art of record does not teach neither singly nor in combination the limitations of claims 1-25.
Edwards (20250086698 ) teaches: associating past auction bidding activity with subsequent financial transactions; receiving, from a web browser plugin of a web browser application executing on a user computing device and via a network, at least a portion of content displayed via a Document Object Model (DOM) of the web browser executing on the user computing device, wherein the content is associated with an auction website: process the content displayed via the DOM to determine bid information indicating a first bid placed by a user on a first auction on the auction website, wherein the bid information comprises: an auction end time, a bid amount, and one or more item identifiers: after the auction end time has elapsed, determine, via the auction website, auction information that indicates a winning bid amount for the first auction: receive, from a transactions history database and based on the auction end time, a plurality of different candidate financial transactions conducted by a financial account associated with the user: receive account data, associated with the financial account, that indicates a location associated with the user: determine an estimated payment amount based on the bid amount and predicted shipping costs associated with the location: select, based on the winning bid amount and based on comparing the estimated payment amount to each transaction amount of the plurality of different candidate financial transactions, a first financial transaction of the plurality of different candidate financial transactions; and output, in a user interface, an association between the first bid and the first financial transaction. However, it lacks the combination of claimed elements of the pending independent claims.
When taken as a whole, the claims are not rendered obvious as the available prior art does not
suggest or otherwise render obvious the noted features nor does the available prior art suggest or otherwise render obvious further modification of the evidence at hand. Such modifications would require substantial reconstruction relying solely on improper hindsight bias, and thus would not be obvious.
Response to Arguments
Applicant’s arguments have been fully considered; Applicant argues with substance:
Applicant respectfully traverses the rejection of claims under § 101. The Examiner has alleged that the claims are directed to a judicial exception without significantly more, that of an abstract idea of a "commercial interaction/fundamental economic practice." Office Action, pp. 3- 8. For the reasons set forth herein, Applicant submits that the claims, at least as amended, are patent eligible. As an initial matter, the claims recite statutory subject matter. Claims 1, 13, 19, and 23 each recite a computer-implemented method, which falls within the "process" category of § 101. This is not in dispute.
1As Applicant's remarks with respect to the Examiner's rejections are sufficient to overcome the Office's allegations set forth in the Office Action, Applicant's silence as to certain assertions or requirements applicable to such allegations (e.g., whether a restriction or election requirement is correct, whether a reference constitutes prior art, motivation to combine references, etc.) does not constitute an acquiescence by Applicant that such allegations are accurate or that such requirements have been met, and Applicant reserves the right to further analyze and dispute such allegations in the future.
Under Step 2A, Prong Two of the Alice/Mayo framework, even assuming arguendo that the claims recite an abstract idea, the claims as amended integrate any such idea into a practical application. Independent claims 1, 13, and 19 now recite a specific, non-generic technical arrangement that improves the functioning of the computing systems and the technological process of generating and serving creative objects, such as digital advertising objects in real time. For example, claim 1 recites "selecting a first neural-network computing system from the plurality of available neural-network computing systems based on the network performance data and the text engine performance data" and "selecting a second neural-network computing system from the plurality of available neural-network computing systems based on the network performance data and the imagery engine performance data". This allows for, among other things, independent processing of text generation and imagery generation and selections according to measured performance characteristics. Claim 1 also recites iteratively providing reconfiguration signals that, "in response to the ad-copy object providing the text and imagery detail outside the quality score range," provides "a first reconfiguration signal to the first neural- network computing system and a first set of reference data" and "a second reconfiguration signal to the second neural-network computing system and a second set of reference data," and iteratively reconfigures both systems "for use in generating subsequent ad-copy objects with text and imagery detail within the quality score range."
Claim 13 recites the similar elements: generating "(a) a text object generated by applying a first prompt dataset to a selected text engine selected from the plurality of text engines based on the network performance data and the text engine performance data, and (b) an imagery object generated by applying a second prompt dataset to a selected imagery engine selected from the plurality of imagery engines based on the network performance data and the imagery engine performance data"; reciting that "the plurality of available neural-network computing systems includes a first neural-network computing system and a second neural-network computing system"; and reciting an iterative reconfiguration loop operating "in response to the one or more creative media objects providing the text and imagery detail outside the quality score range" that reconfigures both systems "for use in generating subsequent ad-copy objects with text and imagery detail within the quality score range."
Claim 19 likewise recites generating "(a) a text object generated by applying a first prompt dataset to a selected text engine selected from a plurality of text engines based on network performance data and text engine performance data, and (b) an imagery object generated by applying a second prompt dataset to a selected imagery engine selected from a plurality of imagery engines based on the network performance data and imagery engine performance data"; reciting that "the plurality of available neural-network computing systems includes a first neural- network computing system and a second neural-network computing system"; and reciting an iterative reconfiguration loop operating "in response to the one or more creative media objects providing the text and imagery detail outside the quality score range" that reconfigures both systems "for use in generating subsequent ad-copy objects with text and imagery detail within the quality score range." This ordered combination of limitations is not merely organizing a commercial interaction on generic computers. Rather, it reflects a particular technical solution to a technical problem: generating high-quality advertising creative objects within the stringent timing constraints of a live advertising auction. The specification explains that the system must operate within a real-time media auction cycle, accommodating network round-trip time, media assembly, and formatting overhead. To meet this latency budget, the system might employ GPU-accelerated inference clusters, a lightweight language model of approximately three billion parameters quantized to 4-bit weights generating approximately 40 tokens in under 30 milliseconds, and an image synthesis model achieving render times on the order of 40-50 milliseconds. The ad-evaluation engine might determine that when the generated content falls below a threshold, that triggers a feedback signal prompting the generation module to increase the score in the next iteration. This iterative adjustment loop can communicate ad quality back to the text engine and imagery engine to improve the quality of the final ad served. Independent claim 23 similarly integrates any alleged abstract idea into a practical
application. Claim 23, as amended, recites receiving "an auction window timing constraint" that "constrains a time period" within which the first media object must be generated, evaluated by "an ad evaluation engine," iteratively modified, and the "one or more generative machine- learning models" are themselves "modif[ied], within the time period." This is a specific technical process for improving real-time content generation within a bounded latency window, not a mere field-of-use limitation.
The Examiner has alleged that the additional elements "do no more than apply or link the use of the recited judicial exception to a particular technological environment/field of use" and do not "improve the functioning of the computing device or another technology/technical field." Office Action, pp. 4-8. Applicant respectfully disagrees. The claims do not merely invoke neural networks or generative models as tools to automate an otherwise-abstract process. Instead, the claims recite a particular, non-generic manner of achieving the result: performance-based selection of computing systems from a plurality of available systems, combined with iterative reconfiguration signals and reference data that cause those systems to reconfigure until the generated content meets a quality score range. This is a technological improvement to the process of real-time ad-creative generation, not a mere instruction to "apply" an abstract idea on a computer.
Even if the claims were deemed not to integrate the alleged abstract idea at Step 2A, the claims satisfy Step 2B because the additional elements, in their ordered combination, amount to significantly more than the abstract idea. The combination of performance-based computing- system selection split over two neural-network computing systems and their corresponding sets of reference data, iterative reconfiguration signaling, and quality-score-range evaluation is not well-understood, routine, or conventional. Indeed, the Examiner's own finding confirms this: "[t]he prior art of record does not teach neither singly nor in combination the limitations of claims 1-25." Office Action, p. 8. If the claimed combination were routine or conventional, one would expect it to appear in the prior art. The Examiner's acknowledgment that it does not supports the conclusion that the claims recite significantly more than an abstract idea. The dependent claims further narrow the independent claims and add additional technical features, such as integration with a real-time programmatic ad-buying platform (claim 2), a generative machine-learning model employing weights of eight bits or fewer to obtain the ad- copy object in fewer processing cycles (claim 9), and an embedding-based language model for determining semantic quality metrics (claim 24), and are patent eligible at least by virtue of their dependency on eligible independent claims.
The pending claims do recite an abstract idea, and the additional elements do not, alone or in combination, integrate the recited abstract idea into a practical application nor do they represent significantly more than the abstract idea itself, as noted above. Generating advertising content to be inserted into an ad slot during a user session represents a business practice/goal, not other technology/technical field; thus, improving this practice pertains to a business practice optimization, not to an improvement to other technology/technical field. As noted above, the pending claimed limitations pertaining to using machine learning/neural network to analyze/generate data, as well as modifying the machine learning models based on criteria/training the machine learning models, do no more than apply or link the use of the recited judicial exception to a particular technological environment/field of use. The pending claims, when implemented, do not effect an improvement to the field of machine learning, nor do they present a specific technical solution- i.e. the pending claims do not recite a novel machine learning architecture, or improved training method; simply using and training machine learning models on data does not transform an abstract idea into a patent eligible invention. There is no technical evidence/technical support in the Spec. that the claimed invention, when implemented, improves the functioning of the computing device itself or other technology/technical field. See Office Action above for the detailed, reasoned 35 USC 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 extension fee 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 ALEXANDRU CIRNU whose telephone number is (571)272-7775. The examiner can normally be reached on M-F 9:00am-5pm. If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, Ilana Spar can be reached on (571) 270-7537. The fax phone number for the organization where this application or proceeding is assigned is 571- 273-8300. Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
Sincerely,
/Alexandru Cirnu/
Primary Patent Examiner, Art Unit 3622
8/11/2026
Prosecution Insights