DETAILED ACTION
Status of the Application
In response filed on July 23, 2026, the Applicant amended claims 1, 19, and 20; added claims 21 and 22; and cancelled claims 13 and 14. Claims 1-12 and 15-22 are pending and currently under consideration for patentability.
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 .
Response to Amendments and Arguments
v Applicant has amended the claims to correct informalities identified in the previous action. These objections have been withdrawn accordingly.
v Applicant’s arguments, with respect to the rejection of claims 1-12 and 15-22 under 35 U.S.C. 101 have been fully considered and are not persuasive. The rejections of claims 1-12 and 15-22 under 35 U.S.C. 101 have been maintained accordingly.
Applicant specifically argues that
1) “The amended claims are not directed to a generic advertising concept or an abstract method of organizing human activity. Claim 1 recites a specific, computer-implemented, two-model technological pipeline. The claim requires processing a first static content item and a user profile using a machine-learned model to generate a new content item by determining a plurality of generated assets. The claim further requires a localized algorithmic evaluation, utilizing a separate machine-learned ranking model to rank these generated assets based on an estimated performance.
The Examiner alleges that optimizing content items to increase user engagement is a "non- technical business advantage/improvement". However, the claimed invention does not broadly preempt targeted advertising. Instead, it provides a highly specific technological solution for digital asset generation. As described in the specification, the estimated performance ranking relies on discrete, measurable digital interactions, such as the granular predicted likelihood of a user interacting with a corresponding content item to execute a hyperlink embedded in the content item. This is not a human mental process or a generic business practice; it is a specific algorithmic data- processing pipeline that transforms a single static asset into an array of tailored, digitally executable candidate assets, which are then algorithmically filtered via a specialized ranking model. For at least this reason, Applicant submits that claim 1 is patent eligible under Step 2A..”
Examiner respectfully disagrees with Applicant’s first argument.
The limitations of a claimed invention need not “broadly preempt targeted advertising” for them to amount to an advertising, marketing or sales activity or behavior, thereby falling within the “certain methods of organizing human activity” subject matter grouping of abstract ideas. The instant claims set-forth a process for generating a new personalized media content item, which may be a message/content for a targeted communication campaign having an objective of having end users engage/click with/on the content item (i.e., an advertising/marketing/communication campaign). One cannot reasonably deny that the claims set for a process that amount to an advertising, marketing or sales activity or behavior, which one is of “[c]ertain methods of organizing human activity” that is an abstract idea. See 2019 Revised Guidance, 84 Fed. Reg. at 52.
Characterizing the steps of this process as a “pipeline” make the limitations of that process no more technical in nature. That the claims require the steps of the process to be carried out using one or more general-purpose computers amounts to mere instructions to implement the abstract idea on a generic computer. The same is true for the requirement to use generic “machine-learned model(s)” to identify the asset characteristics, generate the new content item, and rank the plurality of generated assets, regardless of how many different unspecified models are being used. The requirement to implement the steps of the process using one or more general-purpose computers and with generic “machine-learned model(s)” does not amount to a “highly specific technological solution”. Examiner notes the subject matter from the specification mention by Applicant is not recited in the claims. Even if it were, a requirement for the generated content item to be a digital (e.g., video, image-based) content item that may be served to a computer and predicted to receive clicks (as opposed to a physical advertisement predicted to provoke some other form of engagement/reaction) serves merely to limit the underlying abstract idea to a particular technical environment or field of use.
Applicant’s conclusory statement that “this is not a human mental process” is not persuasive. This type of activity, aside from any computer-related aspects, includes longstanding conduct that existed well before the advent of computers and the Internet, and could be carried out by a human with pen and paper. See CyberSource Corp. v. Retail Decisions, Inc., 654 F.3d 1366, 1375 (Fed. Cir. 2011 – “That purely mental processes can be unpatentable, even when performed by a computer, was precisely the holding of the Supreme Court in Gottschalk v. Benson.”). Applicant fails to explain why a manual approach to performing these functions could not be performed in the human mind or with paper and pen, even if a less desirable alternative.
Applicant specifically argues that
2) “Furthermore, the dependent claims provide additional limitations that integrate the invention into a practical application and provide an inventive concept under Step 2B.For example, dependent Claim 10 requires generating the new content item by editing the first content item using specific, computer-centric operations including "crop, rotate, infill, recolor, defocus, deblur, denoise, relight". These are highly technical, computational graphical manipulation techniques (such as infill, deblur, and denoise) that cannot be performed in the human mind and practically apply the model's output to specific digital image/video modifications..”
Examiner respectfully disagrees with Applicant’s second argument.
Several of the editing techniques (e.g., crop, rotate, infill, recolor, relight) are not “highly technical, computational graphical manipulation techniques”. These techniques were performed using pen and paper long before computer-implementation. Even if Applicant were able to show that one or more of these techniques were not possible using pen and paper (and narrow the claim scope to encompass only this/these techniques), the step of editing the content is still part of the abstract idea (i.e., the advertising, marketing or sales activity or behavior, which one is of “[c]ertain methods of organizing human activity” that is an abstract idea, regardless of whether or not it amounts to a mental process). Examiner notes that editing the content using at least one of these techniques, even if somehow considered to be an “additional” element for the purpose of the eligibility analysis (which the Examiner maintains it is not), would amount to insignificant post-solution activity. Furthermore, these content editing techniques were each well-understood, routine, and conventional at the effective filing date of the claimed invention.
Applicant specifically argues that
3) “Additionally, dependent Claim 15 recites "receiving, via the user interface, inputs providing corrections... and re-generating, using the machine-learned model, a second content item based on the received inputs". As detailed in the specification, integrating these corrective input signals into the machine-learning generation pipeline allows the system to learn from user corrections, avoiding redundant errors in future asset generation cycles, which specifically acts to "improve latency and decrease processing requirements" within the system. Improving the processing efficiency of a machine-learning generation pipeline is a tangible improvement to the functioning of the computer system itself”
Examiner respectfully disagrees with Applicant’s third argument.
Generic iterative “learning” (i.e., model retraining/updating using feedback, such as human correction) does not amount to a technical improvement, or technical solution to a technical problem. Examiner refers to the decision in Recentive Analytics, where the court explored the inquiry of whether claims that do no more than apply established methods of machine learning to a new data environment are patent eligible. The court held that they are not. Central to the courts determination of eligibility was that “(b)oth sets of patents rely on the use of generic machine learning technology in carrying out the claimed methods for generating event schedules and network maps…The machine learning technology described in the patents is conventional…”. The court further found “that the machine learning model be "iteratively trained" or dynamically adjusted in the Machine Learning Training patents do not represent a technological improvement…[because] Iterative training using selected training material and dynamic adjustments based on real-time changes are incident to the very nature of machine learning”. The court also concluded that “the only thing the claims disclose about the use of machine learning is that machine learning is used in a new environment.” Significantly, the court further noted that “We have also held the application of existing technology to a novel database does not create patent eligibility… Stated differently, patents may be directed to abstract ideas where they disclose the use of an "already available [technology], with [its] already available basic functions, to use as [a] tool[ ] in executing the claimed process." SAP Am., 898 F.3d at 1169-70. We think those cases are equally applicable in the machine learning context.”. This analysis applies similarly to Applicant’s alleged “improvement” argument.
v Applicant’s arguments, with respect to the rejection of amended claims 1, 19, and 20 under 35 U.S.C. §103 have been considered, but are not persuasive.
Applicant argues
“Lohiya utilizes generic predictive modeling to optimize systemic rewards within a closed-loop generation system. Lohiya lacks the granular, localized assessment utilized by the Applicant's pipeline, where a generative model first outputs a plurality of specific candidate assets, which are then routed to a distinct, specialized machine-learned ranking model that algorithmically evaluates the assets against a localized digital metric (e.g., predicted likelihood of a user executing a hyperlink).”
Examiner respectfully disagrees. Applicant’s argument is replete with statements that are incommensurate with what is actually claimed. The instant claims do not require “granular, localized assessment” (it is unclear what is meant by this), do not require use of a generative model, do not require first outputting a plurality of specific candidate assets (e.g., merely requires determining them, and the determination and ranking could be performed at the same time), do not require a distinct ranking model (e.g., the machine learned model that determines the plurality of generated assets could also rank them), and the ranking does not require algorithmically evaluating the assets against a localized digital metric (e.g., the ranking is merely required to be based on an estimated performance), and do not require newly generated candidate assets (the could have been previously generated, and the ML model is merely determining which ones to use).
Lohiya teaches determining, using the machine-learned model, a plurality of generated assets, ([0104] “generated content…includes multiple content designs…multiple options for the same audience segment…” – may generate a plurality of new content items with associated generated assets tailored to the first user profile), wherein the machine-learned model is configured to identify asset characteristics associated with the first user profile, and wherein the new content item is generated using the plurality of generated assets ([0053] “the AI/ML architecture provides performance objective prediction feedback for specific items of content, such as a prediction score, that provides…an understanding of factors that make content more likely to elicit a desired response for a specific audience segment… performance objective prediction…for their particular item of content, audience segment, and performance objective, which allows developers to makes changes to the content (or content generation prompt) to generate a new (higher) score to increase the potential for achieving a performance objective of the content before distribution” – therefore the ML models identifies the content factors that associated with increasing engagement/outcome probability with the first user profile (i.e., asset characteristics associated with the first user profile) and uses these to generate the plurality of content item candidates, [0056] “performance information and tuned specifically to generate performance predictions based on the elements of the content and the audience segment accessing the content… learns to generate content to achieve performance objectives for a specific audience segment”, [0067] “content elements, such as the header, subject, body, call to action”, [0096]-[0098] elements…that remain coherent…may include different text, graphics, calls to action, or other elements determined to be optimized for eliciting the requested performance objective KP”, [0102] “content…for different audience segments 904a, 904n can include different content elements, even for the same subject, such as a title or subject line, body text, calls to action, pre-headers, offers, language styles, and/or the like.”). Lohiya also discloses ranking, using the machine-learned model, the plurality of generated assets by using a machine-learned ranking model to rank assets based on an estimated performance of the assets ([0033] “output of the performance prediction model (e.g., a prediction value or score) is used as an indicator of the quality of the content created by the content generation model for prompting one or more recipient responses to the content (i.e., a probability that the content will meet the specified performance object). Through successive iterations, the content generation model learns to generate content with increasing prediction values (i.e., more likely to meet performance objectives)” & [0089] “the generated content 550 comprises reformulated content with a second performance prediction 352 (i.e., higher than the first performance prediction” & [0094]-[0096] “content 560 from the content generation model 140 is also provided to the performance prediction model 250 to determine a performance prediction 352 for the content 450.the differences in the resulting rewards (e.g., the performance prediction 352) can be determined/observed, and the parameters θ can be updated in a direction that returns higher regards (a greater performance prediction 352 for the content 560)…generate content 550 that optimizes for the reward of a higher performance prediction 352” & [0104] “multiple content designs…the content 1050a-n may include various metrics, such as a performance prediction value 1012 and/or performance factors 1014. Accordingly, a user can select among the different content 1050a-n based on the score and/or the content elements…allow a user to visualize the different factors that effected the performance prediction (e.g., increased/decreased the performance prediction), for example, using a logo, offer language, graphics, format, publication source (e.g., email versus YouTube® video), and/or the like.” – therefore the system scores each of the generated candidates/assets using a performance prediction model (i.e., a machine-learned ranking model to rank assets) and ranks them relative to one another (i.e., one is ranked higher when it has a highest estimate probability of a click or other engagement objective)
Applicant argues
“Claim 1 explicitly requires "determining a first user profile that interacts with the plurality of content items based on a relevance score, the relevance score being derived from the user interaction log". The Office Action relies on Naveh to teach this limitation, asserting that Naveh assigns scores indicating probabilities that an image concept will appeal to a user.
This is a mischaracterization of Naveh. Naveh teaches grouping users into cohorts based on predefined shared attributes, and subsequently identifying prevalent image concepts for that pre-established cohort. In Naveh, the correlation score is assigned to the image concept or attribute after the cohort (profile) is already established. Therefore, Naveh does not teach deriving a relevance score from an interaction log and using that score to determine the user profile itself, as explicitly recited in Claim 1.
Examiner respectfully disagrees. It is first noted that Applicant’s specification provides no additional explanation or description of this process. As such, the Examiner’s broadest reasonable interpretation is informed only by the actual claim language itself. The claim language merely requires that the relevance score be derived from the user interaction log, and that the first user profile that interacts with the plurality of content items be somehow determined based on the relevance score. For example, there is no requirement that cohorts/profiles be generated after the generation of a correlation/relevance score. There could be preexisting cohorts/profiles (e.g., demographic segments) and the system generates correlation/relevance score between these cohorts/profiles and image concepts based on the user interaction log (i.e., the system determines the cohort/profile that likes to and/or is likely to like to interact with the plurality of content items based on the relevance score. This embodiment is disclosed by Naveh. Examiner further notes that Naveh also discloses identifying high affinity user profiles (irrespective of cohort) based on probabilities of interest (i.e., relevance scores) that are derived from the user interaction log (see [0042]-[0044]) which may be the interaction log for the target audience group per the combination with Lohiya.
Applicant argues
“Amended Claim 1 further incorporates the features of originally filed dependent Claims 13 and 14, requiring the generation of a plurality of assets based on identified asset characteristics associated with the first user profile, and an algorithmic ranking of those assets using a machine- learned ranking model. To reject these limitations, the Examiner cites Lohiya for teaching that a system generates "multiple content designs" and provides a "performance prediction value". However, Lohiya relies on human intervention rather than an automated machine-learned ranking model. As expressly acknowledged in the Office Action, Lohiya states that "a user can select among the different content... based on the score" using a GUI. Providing an interface for a human user to visualize scores and manually select a content design is not the same as the claimed automated, algorithmic step of "ranking the plurality of generated assets by using a machine-learned ranking model". Furthermore, while Lohiya discusses utilizing prediction scores as internal feedback to mathematically update neural network parameters via reinforcement learning (returning "higher rewards"), this is fundamentally different from deploying a distinct machine-learned ranking model to sort and rank an already-generated plurality of candidate assets to form the final content item. Naveh, which focuses on correlating static image clusters, fails to cure this deficiency. For at least this reason, Applicant submits that Claim 1 is not obvious over the materials cited in the Office Action. Applicant requests withdrawal of the rejection of Claim 1 under § 103”
Examiner respectfully disagrees. Applicant’s argument is incommensurate with what is actually claimed. The instant claims do not require generating a plurality of assets or sorting them. Furthermore, Examiner disagrees that generating/displaying a plurality of scores (using an ML model) based on estimated performance for the plurality of candidates/assets doesn’t amount to ranking the plurality of candidates/assets. The candidates/assets are clearly being ranked, regardless of whether or not a human may subsequently “manually select” a content design.
Applicant argues
“Neither Lohiya nor Naveh teaches or suggests the computational efficiency structures and technical editing manipulations recited in the dependent claims. For example, the prior art combination fails to disclose utilizing the explicit user corrections recited in Claim 15 to re- generate content, which specifically allows the system to avoid redundant errors and reduce processing requirements during subsequent generation cycles. Because the prior art fails to teach all limitations of independent Claim 1, and the dependent claims further distinguish the invention, the § 103 rejections over Lohiya in view of Naveh should be withdrawn”
Examiner respectfully disagrees. The Examiner relied on a third prior art reference
(Serdiukova) to disclose this feature. Applicant has not addressed this reference with respect to this feature
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.
v Claim(s) 1-12 and 15-22 is/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.
Step 1:
Claim(s) 1-12, 15-18, 21, and 22 is/are drawn to methods (i.e., a process), claim(s) 19 is/are drawn to non-transitory media (i.e., a machine/manufacture), and claim(s) 20 is/are drawn to a computing system (i.e., a machine/manufacture). As such, claims 1-12 and 15-22 is/are drawn to one of the statutory categories of invention (Step 1: YES).
Step 2A - Prong One:
In prong one of step 2A, the claim(s) is/are analyzed to evaluate whether it/they recite(s) a judicial exception.
Claim 1 (representative of independent claim(s) 19 and 20) recites/describes the following steps;
obtaining a user interaction log for a target audience group, the target audience group having a plurality of content items that have a similar criteria;
determining a first user profile that interacts with the plurality of content items based on a relevance score, the relevance score being derived from the user interaction log for the target audience group;
obtaining…a first content item from the plurality of content items, the first content item being a static content item;
processing…the first content item and the first user profile to generate the new content item, wherein the new content item is tailored to the first user profile by determining a plurality of generated assets…identify asset characteristics associated with the first user profile, and wherein the new content item is generated using the plurality of generated assets;
ranking the plurality of generated assets…rank assets based on an estimated performance of the assets; and
storing the new content
These steps, under its broadest reasonable interpretation, describe or set-forth a process for generating a new personalized media content item. Specifically, the process involves obtaining a user interaction log for a target audience group, the target audience group having a plurality of content items that have a similar criteria; determining a first user profile that interacts with the plurality of content items based on a relevance score, the relevance score being derived from the user interaction log for the target audience group; obtaining a first content item from the plurality of content items, the first content item being a static content item; processing the first content item and the first user profile to generate the new content item, wherein the new content item is tailored to the first user profile by determining a plurality of generated assets and identifying asset characteristics associated with the first user profile, and wherein the new content item is generated using the plurality of generated assets; ranking the plurality of generated assets based on an estimated performance of the assets; and storing the new content. In light of the original disclosure, it is clear that the new content item may be a message/content for a targeted communication campaign having an objective of having end users engage/click with/on the content item (i.e., an advertising/marketing/communication campaign, see [0003], [0018], [0033], [0046]). This process amounts to a commercial or legal interactions (specifically, an advertising, marketing or sales activity or behavior); and/or managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions). These limitations therefore fall within the “certain methods of organizing human activity” subject matter grouping of abstract ideas.
Additionally, and/or alternatively, each of the above-recited steps/functions, under their broadest reasonable interpretation, encompass a human manually (e.g., in their mind, or using paper and pen) performing one or more concepts performed in the human mind, such as one or more observations, evaluations, judgments, opinions, but for the recitation of generic computer components. If one or more claim limitations, under their broadest reasonable interpretation, covers performance of the limitation(s) in the mind but for the recitation of generic computer components, then it falls within the “mental processes” subject matter grouping of abstract ideas.
As such, the Examiner concludes that claim 1 recites an abstract idea (Step 2A – Prong One: YES).
Independent claim(s) 19 and 20 recite/describe nearly identical steps (and therefore also recite limitations that fall within this subject matter grouping of abstract ideas), and this/these claim(s) is/are therefore determined to recite an abstract idea under the same analysis.
Each of the depending claims likewise recite/describe these steps (by incorporation - and therefore also recite limitations that fall within this subject matter grouping of abstract ideas), and this/these claim(s) is/are therefore determined to recite an abstract idea under the same analysis. Any element(s) recited in a dependent claim that are not specifically identified/addressed by the Examiner under step 2A (prong two) or step 2B of this analysis shall be understood to be an additional part of the abstract idea recited by that particular claim. The same reasoning is similarly applicable to the limitations in the remaining dependent claims, and their respective limitations are not reproduced here for the sake of brevity.
Step 2A - Prong Two:
In prong two of step 2A, an evaluation is made whether a claim recites any additional element, or combination of additional elements, that integrate the exception into a practical application of that exception. An “addition element” is an element that is recited in the claim in addition to (beyond) the judicial exception (i.e., an element/limitation that sets forth an abstract idea is not an additional element). The phrase “integration into a practical application” is defined as requiring an additional element or a combination of additional elements in the claim to apply, rely on, or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception, such that it is more than a drafting effort designed to monopolize the exception.
The claim(s) recite the additional elements/limitations of
“computer-implemented… from a content item database…in the content item database” (claim 1)
“one or more non-transitory, computer readable media storing instructions that are executable by one or more processors to cause a computing system to perform operations… from a content item database…in the content item database” (claim 19)
“a computing system for generating a new content item for a video platform, comprising: one or more processors; one or more non-transitory computer-readable media that collectively store… and instructions that, when executed by the one or more processors, cause the computing system to perform operations, the operations comprising… from a content item database…in the content item database” (claim 20)
“for a video platform” (claims 1, 19, and 20)
“using a machine-learned model…wherein the machine-learned model is configured to…using a machine-learned ranking model to…” (claims 1, 19, and 20)
“wherein the new content item is a video that is presented in the video platform” (claim 2)
“wherein the static content item includes an audio asset” (claim 4)
“wherein the static content item includes a video asset” (claim 6)
“on a user interface…via the user interface” (claim 15)
“using the machine-learned model” (claim 15)
“the user interface comprises” (claim 16)
“using a machine-learned model” (claim 18)
“wherein the machine-learned model…” (claim 22)
The requirement to execute the claimed steps/functions “computer-implemented… from a content item database…in the content item database” (claim 1) or “one or more non-transitory, computer readable media storing instructions that are executable by one or more processors to cause a computing system to perform operations… from a content item database…in the content item database” (claim 19) or “a computing system for generating a new content item for a video platform, comprising: one or more processors; one or more non-transitory computer-readable media that collectively store… and instructions that, when executed by the one or more processors, cause the computing system to perform operations, the operations comprising… from a content item database…in the content item database” (claim 20) and/or “on a user interface…via the user interface” (claim 15) and/or “the user interface comprises” (claim 16) is equivalent to adding the words “apply it” on a generic computer and/or mere instructions to implement the abstract idea on a generic computer. Applicant’s own disclosure explains that these “additional” elements may be embodied as a general-purpose computer (e.g., the published specification at paragraphs [0082]-[0083] “example computing system that can perform…embodiments…The computing device 2 can be any type of computing device, such as, for example, a personal computing device (e.g., laptop or desktop), a mobile computing device (e.g., smartphone or tablet)…or any other type of computing device…processors 12 can be any suitable processing device…”, [0101] “The server computing system 30 can include one or more processors 32 and a memory 34. The one or more processors 32 can be any suitable processing device…”, and [0108]-[0109] “can be implemented in hardware, firmware, or software controlling a general-purpose processor…The network 70 can be any type of communications network, such as a local area network (e.g., intranet), wide area network (e.g., Internet)”). This/these limitation(s) do/does not impose any meaningful limits on practicing the abstract idea, and therefore do/does not integrate the abstract idea into a practical application (see MPEP 2106.05(f)).
The recitation of “using a machine-learned model…wherein the machine-learned model is configured to…using a machine-learned ranking model to…” (claims 1, 19, and 20) and/or “using the machine-learned model” (claim 15) and/or “using a machine-learned model” (claim 18) and/or “wherein the machine-learned model…” (claim 22) provides nothing more than mere instructions to implement an abstract idea on a generic computer. See MPEP 2106.05(f) and the July 2024 Subject Matter Eligibility Examples and corresponding analysis. MPEP 2106.05(f) provides the following considerations for determining whether a claim simply recites a judicial exception with the words “apply it” (or an equivalent), such as mere instructions to implement an abstract idea on a computer: (1) whether the claim recites only the idea of a solution or outcome i.e., the claim fails to recite details of how a solution to a problem is accomplished; (2) whether the claim invokes computers or other machinery merely as a tool to perform an existing process; and (3) the particularity or generality of the application of the judicial exception. The machine-learning model is used to generally apply the abstract idea without placing any limits on how the machine-learning model functions. Rather, these limitations only recite the various outcome (e.g., of “to generate the new content item”, “ranking the plurality of generated assets”) and do not include any details about how these functions are accomplished. See MPEP 2106.05(f) and the July 2024 Subject Matter Eligibility Examples and corresponding analysis. This/these limitation(s) do/does not impose any meaningful limits on practicing the abstract idea, and therefore do/does not integrate the abstract idea into a practical application (see MPEP 2106.05(f)).
The recited additional element(s) of “for a video platform” (claims 1, 19, and 20) and/or “wherein the new content item is a video that is presented in the video platform” (claim 2) and/or “wherein the static content item includes an audio asset” (claim 4) and/or “wherein the static content item includes a video asset” (claim 6) and/or “on a user interface…via the user interface” (claim 15) and/or “the user interface comprises” (claim 16) serves merely to generally link the use of the judicial exception to a particular technological environment or field of use. Specifically, it/they serve(s) to limit the application of the abstract idea to computing environments, such as distributed computing environments and/or the internet, where information is represented digitally, exchanged between computers over a network, and presented using graphical user interfaces. Furthermore, these additional elements serve merely to limit the use of the judicial exception to digital content items and/or specific types of digital content items (e.g., video advertising content, audio advertising content) as opposed to physical forms of content/advertising. This reasoning was demonstrated in Bilski, where it was determined that certain claim elements limiting the basic concept of hedging to commodities and energy markets (merely limiting an abstract idea to one field of use) did not make the concept patentable. This reasoning was demonstrated in Intellectual Ventures I LLC v. Capital One Bank (Fed. Cir. 2015), where the court determined "an abstract idea does not become nonabstract by limiting the invention to a particular field of use or technological environment, such as the Internet [or] a computer"). This/these limitation(s) do/does not impose any meaningful limits on practicing the abstract idea, and therefore do/does not integrate the abstract idea into a practical application (see MPEP 2106.05(g)).
The recitation of “using a machine-learned model…wherein the machine-learned model is configured to…using a machine-learned ranking model to…” (claims 1, 19, and 20) and/or “using the machine-learned model” (claim 15) and/or “using a machine-learned model” (claim 18) and/or “wherein the machine-learned model…” (claim 22) also merely indicates a field of use or technological environment in which the judicial exception is performed. Although the additional element limits the identified judicial exceptions to use of “a machine-learned model”, this type of limitation merely confines the use of the abstract idea to a particular technological environment (machine-learning models) and thus fails to add an inventive concept to the claims. See MPEP 2106.05(h) and the July 2024 Subject Matter Eligibility Examples and corresponding analysis. This/these limitation(s) do/does not impose any meaningful limits on practicing the abstract idea, and therefore do/does not integrate the abstract idea into a practical application (see MPEP 2106.05(g)).
The recited element(s) of “obtaining a user interaction log for a target audience group” (claims 1, 19, and 20) and/or “storing the new content” (claims 1, 19, and 20), even if considered to be an “additional” element for the purpose of the eligibility analysis, would simply append insignificant extra-solution activity to the judicial exception, (e.g., mere pre-solution activity, such as data gathering, in conjunction with an abstract idea; mere post-solution activity in conjunction with an abstract idea). The term “extra-solution activity” is understood as activities incidental to the primary process or product that are merely a nominal or tangential addition to the claim. The recited additional element(s) do are deemed “extra-solution” because all uses of the recited judicial exceptions require such data gathering and output, and because such data gathering and solution-outputting/transmission steps have long been held to be insignificant pre/post-solution activity. This/these limitation(s) do/does not impose any meaningful limits on practicing the abstract idea, and therefore do/does not integrate the abstract idea into a practical application (see MPEP 2106.05(h) and (g)).
Furthermore, although the claims recite a specific sequence of computer-implemented functions, and although the specification suggests certain functions may be advantageous for various reasons (e.g., business reasons), the Examiner has determined that the ordered combination of claim elements (i.e., the claims as a whole) are not directed to an improvement to computer functionality/capabilities, an improvement to a computer-related technology or technological environment, and do not amount to a technology-based solution to a technology-based problem. For example, Applicant’s published specification suggests that it is advantageous to implement the claimed business process because doing so can help create content that is personalized and/or otherwise tailored to the preferences of one or more users which can increase user engagement with the content items thereby helping to better achieve desired campaign objectives (see, for example, Applicant’s published disclosure at paragraphs [0018]-[0019], [0037]). These are non-technical business advantages/improvements. At most, the ordered combination of claim elements is directed to a non-technical improvement to an abstract idea itself (e.g., an improved process for generating a new content item).
Dependent claims 3, 5, 7-12, and 17 fail to include any additional elements. In other words, each of the limitations/elements recited in respective dependent claims 3, 5, 7-12, and 17 is/are further part of the abstract idea as identified by the Examiner for each respective dependent claim (i.e. they are part of the abstract idea recited in each respective claim). For example, claim 3 recites “wherein the static content item includes a text asset”. This is an abstract limitation which further sets forth the abstract idea encompassed by claim 3. This limitation is not an “additional element”, and therefore it is not subject to further analysis under Step 2A- Prong Two or Step 2B. The same logic applies to each of the other dependent claims, whose limitations are not being repeated here for the sake of brevity and clarity. With respect to the other dependent claims not specifically listed here - each of the limitations/elements recited in these dependent claims other than those identified as being “additional” elements above (at the beginning of the Prong One analysis), are further part of the abstract idea encompassed by each respective dependent claim (i.e. it should be understood that these limitations are part of the abstract idea recited in each respective claim).
The Examiner has therefore determined that the additional elements, or combination of additional elements, do not integrate the abstract idea into a practical application. Accordingly, the claim(s) is/are directed to an abstract idea (Step 2A – Prong two: NO).
Step 2B:
In step 2B, the claims are analyzed to determine whether any additional element, or combination of additional elements, is/are sufficient to ensure that the claims amount to significantly more than the judicial exception. This analysis is also termed a search for an "inventive concept." An "inventive concept" is furnished by an element or combination of elements that is recited in the claim in addition to (beyond) the judicial exception, and is sufficient to ensure that the claim as a whole amounts to significantly more than the judicial exception itself. Alice Corp., 134 S. Ct. at 2355, 110 USPQ2d at 1981 (citing Mayo, 566 U.S. at 72-73, 101 USPQ2d at 1966)
As discussed above in “Step 2A – Prong 2”, the requirement to execute the claimed steps/functions “computer-implemented… from a content item database…in the content item database” (claim 1) or “one or more non-transitory, computer readable media storing instructions that are executable by one or more processors to cause a computing system to perform operations… from a content item database…in the content item database” (claim 19) or “a computing system for generating a new content item for a video platform, comprising: one or more processors; one or more non-transitory computer-readable media that collectively store… and instructions that, when executed by the one or more processors, cause the computing system to perform operations, the operations comprising… from a content item database…in the content item database” (claim 20) and/or “on a user interface…via the user interface” (claim 15) and/or “the user interface comprises” (claim 16) is equivalent to adding the words “apply it” on a generic computer and/or mere instructions to implement the abstract idea on a generic computer. These limitations therefore do not qualify as “significantly more” (see MPEP 2106.05(f)).
As discussed above in “Step 2A – Prong 2”, the recitation of “using a machine-learned model…wherein the machine-learned model is configured to…using a machine-learned ranking model to…” (claims 1, 19, and 20) and/or “using the machine-learned model” (claim 15) and/or “using a machine-learned model” (claim 18) and/or “wherein the machine-learned model…” (claim 22) is equivalent to adding the words “apply it” on a generic computer and/or mere instructions to implement the abstract idea on a generic computer. These limitations therefore do not qualify as “significantly more” (see MPEP 2106.05(f)).
As discussed above in “Step 2A – Prong 2”, the recited additional element(s) of “for a video platform” (claims 1, 19, and 20) and/or “wherein the new content item is a video that is presented in the video platform” (claim 2) and/or “wherein the static content item includes an audio asset” (claim 4) and/or “wherein the static content item includes a video asset” (claim 6) and/or “on a user interface…via the user interface” (claim 15) and/or “the user interface comprises” (claim 16) serves merely to generally link the use of the judicial exception to a particular technological environment or field of use. These limitations therefore do not qualify as “significantly more” (see MPEP 2106.05(g)).
As discussed above in “Step 2A – Prong 2”, the recitation of “using a machine-learned model…wherein the machine-learned model is configured to…using a machine-learned ranking model to…” (claims 1, 19, and 20) and/or “using the machine-learned model” (claim 15) and/or “using a machine-learned model” (claim 18) and/or “wherein the machine-learned model…” (claim 22) also serves merely to generally link the use of the judicial exception to a particular technological environment or field of use. These limitations therefore do not qualify as “significantly more” (see MPEP 2106.05(g)).
As discussed above in “Step 2A – Prong 2”, the recited element(s) of “obtaining a user interaction log for a target audience group” (claims 1, 19, and 20) and/or “storing the new content” (claims 1, 19, and 20), even if considered to be an “additional” element for the purpose of the eligibility analysis, would, simply append insignificant extra-solution activity to the judicial exception, (e.g., mere pre-solution activity, such as data gathering, in conjunction with an abstract idea; mere post-solution activity in conjunction with an abstract idea). These additional element(s), taken individually or in combination, additionally amount to well-understood, routine and conventional activities previously known to the industry, specified at a high level of generality, appended to the judicial exception. These additional elements, taken individually or in combination, are well-understood, routine and conventional to those in the field of marketing/advertising. These limitations therefore do not qualify as “significantly more”. (see MPEP 2106.05(d)). This conclusion is based on a factual determination. The determination that associating/storing data in a database is well-understood, routine, and conventional is supported by the 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, 115 USPQ2d at 1092-93), and MPEP 2106.05(d)(II), which note the well-understood, routine, conventional nature of associating/storing data in a database. The determination that receiving data/messages over a network is well-understood, routine, and conventional is supported by Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362; TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016); OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014), and MPEP 2106.05(d)(II), which note the well-understood, routine, conventional nature of receiving data/messages over a network. Furthermore, Examiner takes Official Notice that these steps were well-understood, routine, and conventional at the effective filing date of the claimed invention. Furthermore, the lack of technical detail/description in Applicant’s own specification provides implicit evidence that these steps were well-understood, routine, and conventional.
Viewing the additional limitations in combination also shows that they fail to ensure the claims amount to significantly more than the abstract idea. When considered as an ordered combination, the additional components of the claims add nothing that is not already present when considered separately, and thus simply append the abstract idea with words equivalent to “apply it” on a generic computer and/or mere instructions to implement the abstract idea on a generic computer, generally link the abstract idea to a particular technological environment or field of use, append the abstract idea with insignificant extra solution activity associated with the implementation of the judicial exception, (e.g., mere data gathering, post-solution activity), and appended with well-understood, routine and conventional activities previously known to the industry.
Dependent claims 3, 5, 7-12, and 17 fail to include any additional elements. In other words, each of the limitations/elements recited in respective dependent claims 3, 5, 7-12, and 17 is/are further part of the abstract idea as identified by the Examiner for each respective dependent claim (i.e. they are part of the abstract idea identified by the Examiner to which each respective claim is directed).
The Examiner has therefore determined that no additional element, or combination of additional claims elements is/are sufficient to ensure the claim(s) amount to significantly more than the abstract idea identified above (Step 2B: NO).
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 of this title, 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.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
v Claims 1-11, 17-20, and 22 are rejected under 35 U.S.C. 103 as being unpatentable over Lohiya et al. (U.S. PG Pub No. 2025/0225375, July 10, 2025 - hereinafter "Lahiya”) in view of Naveh et al. (U.S. PG Pub No. 2020/0327375, October 15, 2020 - hereinafter "Naveh”)
With respect to claims 1, 19, and 20, Lohiya teaches a computer-implemented method for generating a new content item for a video platform ([0038] “for video content distributed via a video sharing site, such as YouTube®”); one or more non-transitory, computer readable media storing instructions that are executable by one or more processors to cause a computing system to perform operations ([0107] “implemented as instructions stored on a non-transitory computer-readable storage medium”; and computing system for generating a new content item for a video platform ([0107] “the storage medium 1422 and the processing circuitry 1418 may form a system”), comprising:
one or more processors; (claim 20) ([0162] “computer executable instructions with which one or more processing devices or processing circuitry can execute”, [0206] “executed by one or more processors”)
one or more non-transitory computer-readable media that collectively store a machine-learned model, wherein the machine-learned model is configured to generate the new content item; and (claim 20) ([0162] “the computer-readable storage medium 1802 stores computer executable instructions with which one or more processing devices or processing circuitry can execute.”, Fig 1 tag 140 “content generation model”, [0096] “trained and tuned content generation model”, [0081] “generative language model”)
instructions that, when executed by the one or more processors, cause the computing system to perform operations, the operations comprising: (claim 20) (([0162] “computer executable instructions with which one or more processing devices or processing circuitry can execute”)
obtaining a user interaction log for a target audience group ([0041]-[0042] “leverage content interaction data…for specific audience segments…targeted audience…historical interaction data across different modalities (e.g., numeric data, such as quantifiable content interactions (clicks, reads, interaction duration, etc.), product interactions, product downloads; text data, such as segment labels, surveys and survey interactions; categorical or demographic data, such as geography, funnel stage, age, experience, education, income level, gender, and/or the like)” – obtains user interaction data (i.e., user interaction log) for users of different target audience groups, [0029] “learn how different audience segments interact with different configurations of content….includes results of real-world interactions of multiple audiences with different configurations of content that were developed to elicit certain objectives with the content (e.g., reading the content, clicking on a link in the content, and/or the like). For example, a first marketing email for a product with a first subject line is read by a significant number of members of a first audience segment (e.g., professional programmers, aged 25-35 years old), while being largely ignored by members of a second audience segment (e.g., novice programmers, aged 25-35 years old)” – the interaction data may be for one or more target audience segments for a campaign (i.e., a target audience group), [0047] “As used herein, an “audience segment” is a segment, division, group, or other collection of individuals intended to receive or access content. An audience segment can be defined or divided on various characteristics, including, without limitation, age, gender, income, education, occupation, experience level, exposure (e.g., to content, a product, and/or the like), associated devices, software downloads, associated content consumption mediums and/or platforms”, see also [0062]-[0063] & [0065]-[0067] for discussion of receiving historical interaction data (e.g., click rate, read rate, interaction duration, completion rate) for different target audience groups, [0088] and [0098] discusses input of target audiences (segment definition))
the target audience group having a plurality of content items that have a similar criteria; [0029] “learn how different audience segments interact with different configurations of content….includes results of real-world interactions of multiple audiences with different configurations of content that were developed to elicit certain objectives with the content (e.g., reading the content, clicking on a link in the content, and/or the like). For example, a first marketing email for a product with a first subject line is read by a significant number of members of a first audience segment (e.g., professional programmers, aged 25-35 years old), while being largely ignored by members of a second audience segment (e.g., novice programmers, aged 25-35 years old)…a second marketing email…” – the first target audience segment for a campaign (i.e., a target audience group) has a plurality of content items (e.g., campaign messages, the assets/elements of the messages of the campaign, etc.) that have similar criteria (e.g., targeting criteria because they were part of a campaign targeting certain user segments, performance objective criteria such as reading/clicking, etc.), [0062]-[0065] the target audience interacted with a plurality of content items (e.g., the content that was sent to them) that have similar criteria (e.g., interaction objectives, targeting/selection criteria, content type criteria/restrictions, context, etc.), [0067] “data 210 can include 100 email campaigns, each of which includes a set of specific features (e.g., about 15 features), such as content elements, intent, linked products, subject lines, and/or the like…content of the emails (e.g., content elements, such as the header, subject, body, call to action, time of receipt of email, and/or the like) and/or the audience segment of recipients”)
determining a first user profile that interacts with the plurality of content items ([0058] & [0088]-[0089] in addition to the first content item the system obtains a segment definition that the first content item will be personalized/tailored for using the content generation model and the segment definition may be representative of a particular persona/recipient ([0088] “audience persona” & Fig 9A “Person 1”, [0001] “particular recipient”) and is defined based on associated characteristics ([0088] & [0047] & [0063]) and these characteristics can include age and/or experience and or interests and or location and/or gender and/or spending habits (per [0001] & [0047] & [0058] & [0063] & [0068] & [0088]) and this determined segment definition is the determined “first user profile” that interacts with the plurality of content items, [0041]-[0042] “leverage content interaction data…for specific audience segments…targeted audience…historical interaction data across different modalities (e.g., numeric data, such as quantifiable content interactions (clicks, reads, interaction duration, etc.), product interactions, product downloads; text data, such as segment labels, surveys and survey interactions; categorical or demographic data, such as geography, funnel stage, age, experience, education, income level, gender, and/or the like)” – obtains user interaction data (i.e., user interaction log) for users of different target audience groups, [0029] “learn how different audience segments interact with different configurations of content….includes results of real-world interactions of multiple audiences with different configurations of content that were developed to elicit certain objectives with the content (e.g., reading the content, clicking on a link in the content, and/or the like). For example, a first marketing email for a product with a first subject line is read by a significant number of members of a first audience segment (e.g., professional programmers, aged 25-35 years old), while being largely ignored by members of a second audience segment (e.g., novice programmers, aged 25-35 years old)” – the interaction data may be for one or more target audience segments for a campaign (i.e., a target audience group))
obtaining, from a content item database, a first content item from the plurality of content items, the first content item being a static content item; ([0088]-[0090] “content generation information in the form of various definitions or elements for generating content…includes…content elements (e.g., text, graphics, etc.)…receives content as the input…” – therefore the system obtains a first content item from the plurality of content items (e.g., a content item itself or an element such as an image, headline, call-to-action, etc.) for the ML model to then modify such that it is personalized for a particular target user/audience, [0097] “ the content is optimized to achieve a performance objective, to target an audience segment, and to include content elements (e.g., text, subject, header, body, graphics, etc.) specified in the content generation prompt 410” - this content is “a static content item” as it is capable of being retrieved and served (e.g., for a campaign, on YoutTube, etc.)) without requiring additional server-side processing (Examiner notes that Applicant’s specification provides no definition for “static content item” and therefore may be interpreted broadly in view of the specification, especially considering Applicant’s invention explicitly involves modifying the static content item), per [0126] & [0142] the input from the client may include a “video or image” and may be received from “the data repository 1416” or client/customer profile and/or associated database ( i.e., a content item database), see also [0058] & [0061] for graphics, text, visual content files obtained for input to the content generation model, [0064] & [0067] may be same content as was previously interacted with)
processing, using a machine-learned model, the first content item and the first user profile to generate the new content item, wherein the new content item is tailored to the first user profile; and ([0058] & [0088]-[0089] in addition to the first content item the system obtains a segment definition that the first content item will be personalized/tailored for using the content generation model and the segment definition may be representative of a particular persona/recipient ([0088] “audience personal” & Fig 9A “Person 1”, [0001] “particular recipient”) and is defined based on associated characteristics ([0088] & [0047] & [0063]) and these characteristics can include age and/or experience and or interests and or location and/or gender and/or spending habits (per [0001] & [0047] & [0058] & [0063] & [0068] & [0088]) and this determined segment definition is the “first user profile” that interacts with the plurality of content items, per [0058] “ The content generation module 124 is configured to generate content…that is targeted to a specific…segment definition (e.g., defining the target audience segment)…text or graphics to include in the content…one or more performance objectives desired for content recipients, such as viewing the content, opening an email, responding to an email, clicking on a link, a content interaction duration, and/or the like), a style definition (e.g., information specifying the style, tone, and/or the like of the content, content guidelines (e.g., marketing or corporate style guidelines), generation instructions (e.g., specifying fields of content, such as email subject, call to action, and/or the like, and/or any other information that may be relevant to creating the content), and/or the like” – the content item is personalized/tailored to the first user profile such that probability of a desired engagement/outcome maximized (see [0050]-[0051] “configured to generate content targeted for specific audiences and configured to elicit a specific response from recipients…generating personalized content configured to elicit a specified recipient response… determine features of a specific item of content that directly affect the ability of the item of content to prompt a desired performance objective…language model…to generate content targeted to cause the audience segment to interact with the content and carry out the performance objective…allows a developer to simulate audience responses to content and modify, fine-tune, enhance, and/or the like the content to obtain specific objectives prior to incurring the time and cost of launching the content” - to new content item may be generated by editing/modifying the first content item, see also [0089] & [0096]-[0097]), [0104] “generated content…includes multiple content designs…multiple options for the same audience segment…” – may generate a plurality of new content items tailored to the first user profile)
by determining a plurality of generated assets, ([0104] “generated content…includes multiple content designs…multiple options for the same audience segment…” – may generate a plurality of new content items with associated generated assets tailored to the first user profile)
wherein the machine-learned model is configured to identify asset characteristics associated with the first user profile, and wherein the new content item is generated using the plurality of generated assets ([0053] “the AI/ML architecture provides performance objective prediction feedback for specific items of content, such as a prediction score, that provides…an understanding of factors that make content more likely to elicit a desired response for a specific audience segment… performance objective prediction…for their particular item of content, audience segment, and performance objective, which allows developers to makes changes to the content (or content generation prompt) to generate a new (higher) score to increase the potential for achieving a performance objective of the content before distribution” – therefore the ML models identifies the content factors that associated with increasing engagement/outcome probability with the first user profile (i.e., asset characteristics associated with the first user profile) and uses these to generate the plurality of content item candidates, [0056] “performance information and tuned specifically to generate performance predictions based on the elements of the content and the audience segment accessing the content… learns to generate content to achieve performance objectives for a specific audience segment”, [0067] “content elements, such as the header, subject, body, call to action”, [0096]-[0098] elements…that remain coherent…may include different text, graphics, calls to action, or other elements determined to be optimized for eliciting the requested performance objective KP”, [0102] “content…for different audience segments 904a, 904n can include different content elements, even for the same subject, such as a title or subject line, body text, calls to action, pre-headers, offers, language styles, and/or the like.”)
ranking the plurality of generated assets by using a machine-learned ranking model to rank assets based on an estimated performance of the assets ([0033] “output of the performance prediction model (e.g., a prediction value or score) is used as an indicator of the quality of the content created by the content generation model for prompting one or more recipient responses to the content (i.e., a probability that the content will meet the specified performance object). Through successive iterations, the content generation model learns to generate content with increasing prediction values (i.e., more likely to meet performance objectives)” & [0089] “the generated content 550 comprises reformulated content with a second performance prediction 352 (i.e., higher than the first performance prediction” & [0094]-[0096] “content 560 from the content generation model 140 is also provided to the performance prediction model 250 to determine a performance prediction 352 for the content 450.the differences in the resulting rewards (e.g., the performance prediction 352) can be determined/observed, and the parameters θ can be updated in a direction that returns higher regards (a greater performance prediction 352 for the content 560)…generate content 550 that optimizes for the reward of a higher performance prediction 352” & [0104] “multiple content designs…the content 1050a-n may include various metrics, such as a performance prediction value 1012 and/or performance factors 1014. Accordingly, a user can select among the different content 1050a-n based on the score and/or the content elements…allow a user to visualize the different factors that effected the performance prediction (e.g., increased/decreased the performance prediction), for example, using a logo, offer language, graphics, format, publication source (e.g., email versus YouTube® video), and/or the like.” – therefore the system scores each of the generated candidates/assets using a performance prediction model (i.e., a machine-learned ranking model to rank assets) and ranks them relative to one another (i.e., one is ranked higher when it has a highest estimate probability of a click or other engagement objective)
storing the new content item in the content item database ([0126] & [0142] the output to the client may be sent to the same “the data repository 1416” and/or client associated database ( i.e., a content item database) – Examiner notes Naveh also discloses storing the generated new content item in a content item database so that it may be retrieved later and served to the target user (see [0081]-[0085] “content store…advertisement…one or more images…content for presentation to a user…targeting criteria…” which may be the generated content per [0029] & [0045] & [0057] & [0066] & [0096] & [0101])
Lohiya does not appear to disclose,
that interacts with the plurality of content items based on a relevance score, the relevance score being derived from the user interaction log for the target audience group
However, Naveh discloses a system that generates personalized static content (e.g., advertisements) for a single user profile ([0046] “the image concept modeler module 120 may analyze and model a single user…may be used to generate custom machine learning models for individuals so that content may be customized (automatically as described herein or otherwise) specifically for that individual, [0048] “may predict that an image concept may appeal to a particular user….may generate an output that indicates that a combination of two or more image concepts in an image may increase the probability that the image will be interacted with”, [0066] “automatically generates images using image concepts, according to an example….generate an image having image concepts predicted to appeal to the audience. In some examples, the image generation module 140 may include a Generative Adversarial Network (GAN) that is trained to generate candidate images based on the image concepts identified by the image concept model 210….The generator may, for example, apply image style transfer techniques to automatically (programmatically based on machine learned information such as from the discriminator) generate candidate images. In particular, the generator may apply an image concept to modify an existing image or may combine two image concepts to create a new image”, [0059] “may receive an audience definition from a campaign portal 510. The audience definition may specify an intended audience, including a user or group of users, that is to receive an image…a requester such as an advertiser may specify an audience to receive images relating to an ad campaign to market goods or services….may incorporate the image concepts into images for the ad campaign”). Naveh further discloses
determining a first user profile that interacts with the plurality of content items based on a relevance score, the relevance score being derived from the user interaction log for the target audience group ([0038]-[0045 “may observe that a user interacted with various images based on the user interaction information. Each of the images may include respective sets of image concepts, as defined in the image concept information. Some of the image concepts may have contributed to the user interaction with the images while others may not. The image concept modeler module 120 may generate/train a machine learning model based on image concepts that commonly appear in images interacted with by the user, correlating such image concepts with user interaction of images that contain the image concepts….correlate image concepts 312 (illustrated as image concepts 312A-D) identified from images 310 (illustrated as images 310A-N) interacted with by a first user (User 1). User 1 may be associated with attributes (A-C). The image concept modeler module 120 may identify image concepts 312 (illustrated as image concept “shiny ring” 312A, image concept “cat” 312B, image concept “bird” 312C and image concept “dog” 312D) from the images 310. The image concept modeler module 120 may identify the most prevalent image concepts (based on deviation from a mean or predefined threshold). For example, the image concept modeler module 120 may determine that image concepts 312A and 312B are most prevalent….These two image concepts may be the most prevalent image concepts among all image concepts observed in images 310 interacted with by the first user… may correlate attributes A-C of the first user with the image concepts 312A and 312B. By repeating this procedure over a population of users, the image concept modeler module 120 may generate a correlative pattern of user attributes and image concepts that appear in images interacted with by the users….Because the image concept 312A is common among the most prevalent image concepts for the first user and the second user, and because the first user and the second user share attribute (A), the image concept modeler module 120 may correlate attribute (A) with the image concept 312A….may assign a score for each correlated attribute and image concept. The score may indicate a probability that a given image concept will appeal to a user with a corresponding correlated attribute based on the prevalence of correlation observed in the interaction information. …affinity to image concepts…the image concept modeler module 120 may perform the correlations based on cohorts, or groups, of users that are clustered together based on shared attributes….The image concept modeler module 120 may identify image concepts that are most prevalent in images interacted with by the cohort. As such, the image concept modeler module 120 may correlate the cohort with the most prevalent image concepts. More specifically, the image concept modeler module 120 may correlate attributes shared by the cohort with the most prevalent image concepts….this information may be useful for generating images for an audience that includes the cohort” – therefore the system determines which user profiles and user cohorts interact most frequently with different content items and content item concepts/attributes based on a determined frequency of interaction and/or correlation strength and/or correlation score that is based on the interaction log (i.e., determining a first user profile that interacts with the plurality of content items based on a relevance score (frequency of interaction, correlation strength, correlation score) derived from the user interaction log for a target audience group) and associates these users and concepts/contents together for subsequent optimization and personalizing of the content items, see also [0027] & [0030] & [0085]-[0086])
Naveh suggests it is advantageous to determine a first user profile that interacts with the plurality of content items based on a relevance score, the relevance score being derived from the user interaction log for the target audience group, because doing so can identify correlations between certain users (and/or associated user attributes) and certain content items (and/or associated content concepts/attributes) to identify user profile and content combinations likely to result in desired engagements which can increase the effectiveness of a marketing campaign ([0017] & & [0059] &[0084] & [0096]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method, media, and system of Lohiya to include determining a first user profile that interacts with the plurality of content items based on a relevance score, the relevance score being derived from the user interaction log for the target audience group, as taught by Naveh, because doing so can identify correlations between certain users (and/or associated user attributes) and certain content items (and/or associated content concepts/attributes) to identify user profile and content combinations likely to result in desired engagements which can increase the effectiveness of a marketing campaign.
With respect to claim 2, Lohiya teaches the method of claim 1;
wherein the new content item is a video that is presented in the video platform ([0030]-[0035] “content…text, video, images… the content generation model generates output in the form of content relating to the subject that is directed to cause a recipient of the target audience to perform the performance objective. For example…the content generation model generates a first marketing video with content predicted by the performance prediction model to cause a recipient of the target audience to visit the vendor website” – the content output from the content generation model (the new content item) may be a video item, [0038] “for video content distributed via a video sharing site, such as YouTube®” – may be presented in the video platform, [0061] “delivery form…a video…”)
With respect to claim 3, Lohiya teaches the method of claim 1;
wherein static content item includes a text asset ([0088]-[0090] “content generation information in the form of various definitions or elements for generating content…includes…content elements (e.g., text, graphics, etc.)…receives content as the input…” – therefore the system obtains a first content item (static content item) includes a text asset, [0097] “ the content is optimized to achieve a performance objective, to target an audience segment, and to include content elements (e.g., text, subject, header, body, graphics, etc.) specified in the content generation prompt 410”, [0044] “As used herein, “content” or any variations thereof refers to any type of visual, graphical, textual, auditory, combinations thereof, and/or the like information for presentation to a recipient. Non-limiting examples of content include digital media, any website, any email, any graphic, any video, any image, any audio, any text, any computer program, and/or any other form of information and/or any combinations thereof”, [0030] “receives content (e.g.,…text…”, [0052] “content…email, website”)
With respect to claim 4, Lohiya teaches the method of claim 1;
wherein the static content item includes an audio asset ([0088]-[0090] content (i.e., the static content) is received as the input and per [0044] “ “content” or any variations thereof refers to any type of visual, graphical, textual, auditory, combinations thereof, and/or the like information for presentation to a recipient. Non-limiting examples of content include digital media, any website, any email, any graphic, any video, any image, any audio, any text, any computer program, and/or any other form of information and/or any combinations thereof” – therefore may be an audio asset and/or a video with audio, [0143] audio data)
With respect to claim 5, Lohiya teaches the method of claim 1;
wherein the static content item includes an image asset ([0088]-[0090] “content generation information in the form of various definitions or elements for generating content…includes…content elements (e.g., text, graphics, etc.)…receives content as the input…” – therefore the system obtains a first content item (static content item) includes a text asset, [0097] “ the content is optimized to achieve a performance objective, to target an audience segment, and to include content elements (e.g., text, subject, header, body, graphics, etc.) specified in the content generation prompt 410”, [0030] “receives content (e.g.,…video, images…” [0044] “As used herein, “content” or any variations thereof refers to any type of visual, graphical, textual, auditory, combinations thereof, and/or the like information for presentation to a recipient. Non-limiting examples of content include digital media, any website, any email, any graphic, any video, any image, any audio, any text, any computer program, and/or any other form of information and/or any combinations thereof”)
With respect to claim 6, Lohiya teaches the method of claim 1;
wherein the static content item includes a video asset (([0088]-[0090] content (i.e., the static content) is received as the input and per [0044] ““content” or any variations thereof refers to any type of visual, graphical, textual, auditory, combinations thereof, and/or the like information for presentation to a recipient. Non-limiting examples of content include digital media, any website, any email, any graphic, any video, any image, any audio, any text, any computer program, and/or any other form of information and/or any combinations thereof” – therefore may be a video asset, [0030] “receives content (e.g.,…video, images…”)
With respect to claim 7, Lohiya teaches the method of claim 1;
wherein the static content item includes two or more modalities selected from: text, image, audio, or video ([0088]-[0090] “content generation information in the form of various definitions or elements for generating content…includes…content elements (e.g., text, graphics, etc.)…receives content as the input…” – therefore the system obtains a first content item (static content item) includes a text and image modalities, [0097] “ the content is optimized to achieve a performance objective, to target an audience segment, and to include content elements (e.g., text, subject, header, body, graphics, etc.) specified in the content generation prompt 410” – at least text and image, [0030] “receives content (e.g.,…video, images…” [0044] “As used herein, “content” or any variations thereof refers to any type of visual, graphical, textual, auditory, combinations thereof, and/or the like information for presentation to a recipient. Non-limiting examples of content include digital media, any website, any email, any graphic, any video, any image, any audio, any text, any computer program, and/or any other form of information and/or any combinations thereof”)
With respect to claim 8, Lohiya teaches the method of claim 1;
wherein the static content item is obtained from a content account ([0126] & [0142] the input from the client may include a “video or image” and may be received from “the data repository 1416” which may be associated with a client/customer profile and/or associated database (i.e., a content account), see also [0058] & [0061])
With respect to claim 9, Lohiya teaches the method of claim 8;
wherein the new content item is generated using information derived from an account profile of a client account ([0142] “data sources…customer profiles…product inventories…”) – Examiner notes Naveh also discloses this limitation at ([0059] “campaign portal…requestor…” & [0066] & [0095]-[0097] where the campaign definition information is received from campaign module/portal which is equivalent to an account profile for a client account (advertiser client))
With respect to claim 10, Lohiya teaches the method of claim 1;
comprising: generating the new content item by editing the first content item using at least one of the following editing operations: crop, rotate, infill, recolor, defocus, deblur, denoise, relight ([0050]-[0051] “configured to generate content targeted for specific audiences and configured to elicit a specific response from recipients…generating personalized content configured to elicit a specified recipient response…determine features of a specific item of content that directly affect the ability of the item of content to prompt a desired performance objective…language model…to generate content targeted to cause the audience segment to interact with the content and carry out the performance objective…allows a developer to simulate audience responses to content and modify, fine-tune, enhance, and/or the like the content to obtain specific objectives prior to incurring the time and cost of launching the content” - to new content item may be generated by editing/modifying the first content item, see also [0089] & [0096]-[0097]), [0088] “content generation information in the form of various definitions or elements for generating content…. includes information…such as… a style definition 608 and generation instructions 610. The style definition 608 indicates any stylistic characteristics, such as persuasiveness, corporate guidelines, preferred terms, prohibited terms, colors, content look-and-feel, and/or the like….can also include text length, character limits, token limits, use/non-use of emojis or other symbols, and/or the like…” – therefore editing/modifying the content based on the stylistic characteristics may include editing/adjusting the color (i.e., recolor operation), [0091] “stylistic instructions”)
With respect to claim 11, Lohiya teaches the method of claim 1;
wherein the new content item is generated based on a parameter of the first user profile ([0058] & [0088]-[0089] in addition to the first content item the system obtains a segment definition that the first content item will be personalized/tailored for using the content generation model and the segment definition may be representative of a particular persona/recipient ([0088] “audience personal” & Fig 9A “Person 1”, [0001] “particular recipient”) and is defined based on associated characteristics ([0088] & [0047] & [0063]) and these characteristics can include age and/or experience and or interests and or location and/or gender and/or spending habits (per [0001] & [0047] & [0058] & [0063] & [0068] & [0088]) and each of these characteristics are “a parameter of the first user profile” that are used to generate the content)
With respect to claim 17, Lohiya teaches the method of claim 1;
wherein the new content item comprises two or more categories of the following categories: images, headlines, descriptions, videos, logos, colors, sitelinks, calls to action, audio ([0104] “multiple content designs…different factors that effected the performance prediction (e.g., increased/decreased the performance prediction), for example, using a logo, offer language, graphics, format, publication source (e.g., email versus YouTube® video), and/or the like” – the content comprises content from at least image/logo category and video, [0067] “content elements, such as the header, subject, body, call to action”, [0097] “Content 850a and 850n may include different text, graphics, calls to action, or other elements determined to be optimized for eliciting the requested performance objective KPI” see also [0090])
With respect to claim 18, Lohiya teaches the method of claim 1;
determining a second user profile that interacts with the plurality of content items (- [0058] & [0088]-[0089] in addition to the first content item the system obtains a segment definition that the first content item will be personalized/tailored for using the content generation model and the segment definition may be representative of a particular persona/recipient ([0088] “audience persona” & Fig 9A “Person 2”, and is defined based on associated characteristics ([0088] & [0047] & [0063]) and these characteristics can include age and/or experience and or interests and or location and/or gender and/or spending habits (per [0001] & [0047] & [0058] & [0063] & [0068] & [0088]) and this determined persona/person definition is the determined “second user profile” that interacts with the plurality of content items, [0041]-[0042] “leverage content interaction data…for specific audience segments…targeted audience…historical interaction data across different modalities (e.g., numeric data, such as quantifiable content interactions (clicks, reads, interaction duration, etc.), product interactions, product downloads; text data, such as segment labels, surveys and survey interactions; categorical or demographic data, such as geography, funnel stage, age, experience, education, income level, gender, and/or the like)” – obtains user interaction data (i.e., user interaction log) for users of different target audience groups, [0029] “learn how different audience segments interact with different configurations of content….includes results of real-world interactions of multiple audiences with different configurations of content that were developed to elicit certain objectives with the content (e.g., reading the content, clicking on a link in the content, and/or the like). For example, a first marketing email for a product with a first subject line is read by a significant number of members of a first audience segment (e.g., professional programmers, aged 25-35 years old), while being largely ignored by members of a second audience segment (e.g., novice programmers, aged 25-35 years old)” – Examiner notes that this limitation amounts to repeating the same process a second time which is envisioned by Lohiya)
processing, using a machine-learned model, the first content item and the second user profile to generate a second content item, wherein the second content item is tailored to the second user profile; and ([0058] & [0088]-[0089] in addition to the first content item the system obtains a segment definition that the first content item will be personalized/tailored for using the content generation model and the segment definition may be representative of a particular persona/recipient ([0088] “audience personal” & Fig 9A “Person 2”, [0001] “particular recipient”) and is defined based on associated characteristics ([0088] & [0047] & [0063]) and these characteristics can include age and/or experience and or interests and or location and/or gender and/or spending habits (per [0001] & [0047] & [0058] & [0063] & [0068] & [0088]) and this determined segment definition is the “first user profile” that interacts with the plurality of content items, per [0058] “ The content generation module 124 is configured to generate content…that is targeted to a specific…segment definition (e.g., defining the target audience segment)…text or graphics to include in the content…one or more performance objectives desired for content recipients, such as viewing the content, opening an email, responding to an email, clicking on a link, a content interaction duration, and/or the like), a style definition (e.g., information specifying the style, tone, and/or the like of the content, content guidelines (e.g., marketing or corporate style guidelines), generation instructions (e.g., specifying fields of content, such as email subject, call to action, and/or the like, and/or any other information that may be relevant to creating the content), and/or the like” – the content item is personalized/tailored to the first user profile such that probability of a desired engagement/outcome maximized (see [0050]-[0051] “configured to generate content targeted for specific audiences and configured to elicit a specific response from recipients…generating personalized content configured to elicit a specified recipient response… determine features of a specific item of content that directly affect the ability of the item of content to prompt a desired performance objective…language model…to generate content targeted to cause the audience segment to interact with the content and carry out the performance objective…allows a developer to simulate audience responses to content and modify, fine-tune, enhance, and/or the like the content to obtain specific objectives prior to incurring the time and cost of launching the content” - to new content item may be generated by editing/modifying the first content item, see also [0089] & [0096]-[0097]), [0104] “generated content…includes multiple content designs…multiple options for the same audience segment…” – may generate a plurality of new content items tailored to the second user profile, – Examiner notes that this limitation amounts to repeating the same process a second time which is envisioned by Lohiya)
storing the second content item in the content item database ([0126] & [0142] the output to the client may be sent to the same “the data repository 1416” and/or client associated database ( i.e., a content item database) – Examiner notes Naveh also discloses storing the generated new content item in a content item database so that it may be retrieved later and served to the target user (see [0081]-[0085] “content store…advertisement…one or more images…content for presentation to a user…targeting criteria…” which may be the generated content per [0029] & [0045] & [0057] & [0066] & [0096] & [0101])
Lohiya does not appear to disclose,
that interacts with the plurality of content items based on a relevance score, the relevance score being derived from the user interaction log for the target audience group
However, Naveh further discloses
determining a second user profile that interacts with the plurality of content items based on a relevance score, the relevance score being derived from the user interaction log for the target audience group ([0038]-[0045 “may observe that a user interacted with various images based on the user interaction information. Each of the images may include respective sets of image concepts, as defined in the image concept information. Some of the image concepts may have contributed to the user interaction with the images while others may not. The image concept modeler module 120 may generate/train a machine learning model based on image concepts that commonly appear in images interacted with by the user, correlating such image concepts with user interaction of images that contain the image concepts….correlate image concepts 312 (illustrated as image concepts 312A-D) identified from images 310 (illustrated as images 310A-N) interacted with by a first user (User 1). User 1 may be associated with attributes (A-C). The image concept modeler module 120 may identify image concepts 312 (illustrated as image concept “shiny ring” 312A, image concept “cat” 312B, image concept “bird” 312C and image concept “dog” 312D) from the images 310. The image concept modeler module 120 may identify the most prevalent image concepts (based on deviation from a mean or predefined threshold). For example, the image concept modeler module 120 may determine that image concepts 312A and 312B are most prevalent….These two image concepts may be the most prevalent image concepts among all image concepts observed in images 310 interacted with by the first user… may correlate attributes A-C of the first user with the image concepts 312A and 312B. By repeating this procedure over a population of users, the image concept modeler module 120 may generate a correlative pattern of user attributes and image concepts that appear in images interacted with by the users….Because the image concept 312A is common among the most prevalent image concepts for the first user and the second user, and because the first user and the second user share attribute (A), the image concept modeler module 120 may correlate attribute (A) with the image concept 312A….may assign a score for each correlated attribute and image concept. The score may indicate a probability that a given image concept will appeal to a user with a corresponding correlated attribute based on the prevalence of correlation observed in the interaction information. …affinity to image concepts…the image concept modeler module 120 may perform the correlations based on cohorts, or groups, of users that are clustered together based on shared attributes….The image concept modeler module 120 may identify image concepts that are most prevalent in images interacted with by the cohort. As such, the image concept modeler module 120 may correlate the cohort with the most prevalent image concepts. More specifically, the image concept modeler module 120 may correlate attributes shared by the cohort with the most prevalent image concepts….this information may be useful for generating images for an audience that includes the cohort” – therefore the system determines which user profiles and user cohorts interact most frequently with different content items and content item concepts/attributes based on a determined frequency of interaction and/or correlation strength and/or correlation score that is based on the interaction log (i.e., determining a second user profile that interacts with the plurality of content items based on a relevance score (frequency of interaction, correlation strength, correlation score) derived from the user interaction log for a target audience group) and associates these users and concepts/contents together for subsequent optimization and personalizing of the content items, see also [0027] & [0030] & [0085]-[0086] - Examiner notes that this limitation amounts to repeating the same process a second time which is envisioned by Lohiya))
Naveh suggests it is advantageous to determine a second user profile that interacts with the plurality of content items based on a relevance score, the relevance score being derived from the user interaction log for the target audience group, because doing so can identify correlations between certain users (and/or associated user attributes) and certain content items (and/or associated content concepts/attributes) to identify user profile and content combinations likely to result in desired engagements which can increase the effectiveness of a marketing campaign ([0017] & & [0059] &[0084] & [0096]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method, media, and system of Lohiya to include determining a second user profile that interacts with the plurality of content items based on a relevance score, the relevance score being derived from the user interaction log for the target audience group, as taught by Naveh, because doing so can identify correlations between certain users (and/or associated user attributes) and certain content items (and/or associated content concepts/attributes) to identify user profile and content combinations likely to result in desired engagements which can increase the effectiveness of a marketing campaign.
With respect to claim 22, Lohiya teaches the method of claim 1;
wherein the machine- learned model generates the new content item using an offline pipeline that runs periodically to incorporate new engagement statistics ([0130[ “the data collector 1502 collects data 1512 from one or more data sources to use as training data for the ML model 1230. The data collector 1502 collects different types of data 1512, such as text information, audio information, image information, video information, graphic information, and so forth. The model trainer 1504 receives as input the collected data and uses a portion of the collected data as test data for an AI/ML algorithm to train the ML model 1230. The model evaluator 1506 evaluates and improves the trained ML model 1230 using a portion of the collected data as test data to test the ML model 1230. The model evaluator 1506 also uses feedback information from the deployed ML model 1230. The model inferencer 1508 implements the trained ML model 1230 to receive as input new unseen data, generate one or more inferences on the new data, and output a result such as an alert, a recommendation or other post-solution activity.” and [0134] “a pipeline that can take in raw data, process it, train an ML model 1430, evaluate performance of the trained ML model 1430, and deploy the tested ML model 1430 as the trained ML model 1430 in a production environment, and continuously monitor and maintain it.” And [0146]-[0149] “The evaluation process includes providing feedback 1618 to the model trainer 1504. The model trainer 1504 re-trains the ML model 1430 to improve performance in an iterative manner. The model evaluator 1506 is communicatively coupled to a model inferencer 1508. The model inferencer 1508 provides AI/ML model inference output (e.g., inferences, predictions or decisions). Once the ML model 1430 is trained and evaluated, it is deployed in a production environment where it is used to make predictions on new data. The model inferencer 1508 receives the evaluated ML model 1430 as input 1614. The model inferencer 1508 uses the evaluated ML model 1430 to produce insights or predictions on real data, which is deployed as a final production ML model 1430. The inference output of the ML model 1430 is use case specific. The model inferencer 1508 also performs model monitoring and maintenance, which involves continuously monitoring performance of the ML model 1430 in the production environment and making any necessary updates or modifications to maintain its accuracy and effectiveness. The model inferencer 1508 provides feedback 1618 to the data collector 1502 to train or re-train the ML model 1430. The feedback 1618 includes model performance feedback information, which is used for monitoring and improving performance of the ML model 1430” – therefore the system periodically updates/re-trains the ML model (e.g., based on new engagement statistics) and this ML model is used to generate new content items and therefore the machine- learned model generates the new content item using an offline pipeline that runs periodically to incorporate new engagement statistics)
v Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over Lohiya in view of Naveh, as applied to claim 1 above, and further in view of Seth (U.S. Patent No. 12,430,388 September, 30 2025 - hereinafter "Seth”)
With respect to claim 12, Lohiya teaches the method of claim 1;
wherein the new content item is generated based on a set of content item guidelines for generating content items using a pre-existing image asset, ([0088] “content generation information in the form of various definitions or elements for generating content…. includes information…such as…content elements (e.g., text, graphics, etc.)….can include instructions defining how the content is to be delivered, such as email, text, video, website, and/or the like….can include additional elements, such as a style definition 608 and generation instructions 610. The style definition 608 indicates any stylistic characteristics, such as persuasiveness, corporate guidelines, preferred terms, prohibited terms, colors, content look-and-feel, and/or the like….can also include text length, character limits, token limits, use/non-use of emojis or other symbols, and/or the like…” – therefore the new content item is generated based on a set of content item guidelines for generating content items (e.g., required delivery medium, requires stylistic characteristics, coloring, etc.) and required pre-existing image asset), [0091] “stylistic instructions”)
Lohiya does not appear to disclose,
the set of content item guidelines include resolution specifications, aspect ratio specifications, or orientation specifications
However, Seth discloses
the set of content item guidelines include resolution specifications, aspect ratio specifications, or orientation specifications (10:46-54 “user feedback on how to adjust the transformed design content item, such as font being too small, the audio being too soft, subject/background/object size too big/small, resolution too high/low, colors too bright/dark…”, 11:45-63 “converts the user data to the format directly processible by the generative model(s)…can be considered in adjusting the personalized design content items rejected by the user, such as bigger/smaller subject/background/object size, higher/lower resolution, brighter/darker colors”)
Seth suggests it is advantageous to include wherein the set of content item guidelines include resolution specifications, aspect ratio specifications, or orientation specifications, because doing so can ensure the new content item is generated based on the specific requirements desired by the content creator and these are comment specification that may be important to a content creator (10:46-54, 11:45-63).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Lohiya in view of Naveh to include wherein the set of content item guidelines include resolution specifications, aspect ratio specifications, or orientation specifications, as taught by Seth, because doing so can ensure the new content item is generated based on the specific requirements desired by the content creator and these are comment specification that may be important to a content creator.
v Claims 15, 16, and 21 are rejected under 35 U.S.C. 103 as being unpatentable over Lohiya in view of Naveh, as applied to claim 1 above, and further in view of Serdiukova et al. (U.S. PG Pub No. 2024/0303280 September 12, 2024 - hereinafter "Serdiukova”)
With respect to claim 15, Lohiya teaches the method of claim 1;
comprising: presenting, on a user interface accessible by a client account, the new content item for review; ([0104] “GUI 1030 displays generated content 1050a-n…generated content…includes multiple content designs…multiple options for the same audience segment…the content 1050a-n may include various metrics, such as a performance prediction value 1012 and/or performance factors 1014. Accordingly, a user can select among the different content 1050a-n based on the score and/or the content elements…allow a user to visualize the different factors that effected the performance prediction (e.g., increased/decreased the performance prediction), for example, using a logo, offer language, graphics, format, publication source (e.g., email versus YouTube® video), and/or the like” – presenting the new content item(s) candidate(s) for review on a GUI)
receiving, via the user interface, inputs ([0104] “a user can select among the different content 1050a-n”)
Lohiya does not appear to disclose,
inputs providing corrections to the new content item; and re-generating, using the machine-learned model, a second content item based on the received inputs
However, Serdiukova discloses
inputs providing corrections to the new content item; and re-generating, using the machine-learned model, a second content item based on the received inputs (Fig 5 “other” and [0039] “the user may provide the AI with feedback on the subject lines, for example, by interacting with a thumbs up icon, a thumbs down icon, and/or by providing unstructured feedback in a text prompt. Other forms and mechanisms for providing feedback are also contemplated within the scope of the present disclosure”, [0053]-[0054] “the content generation service may receive feedback associated with the candidate strings…the content generation service described herein may use feedback provided by the cloud clients to update or refine the machine learning algorithms used to generate candidate subject lines, thereby enabling the content generation service to provide more pertinent suggestions in subsequent attempts”, see also [0018] & [0029] “subsequent iterations”)
Serdiukova suggests it is advantageous to include inputs providing corrections to the new content item; and re-generating, using the machine-learned model, a second content item based on the received inputs, because doing so can help ensure the generated content is desirable to the liking of the content creator ([0018] & [0029] & [0039] & [0053]-[0054]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Lohiya in view of Naveh to include inputs providing corrections to the new content item; and re-generating, using the machine-learned model, a second content item based on the received inputs, as taught by Serdiukova, because doing so can help ensure the generated content is desirable to the liking of the content creator.
With respect to claim 16, Lohiya, Naveh, and Serdiukova teach the method of claim 15. Lohiya does not appear to disclose,
wherein the user interface comprises a natural language input element for receiving corrective inputs in natural language format, wherein the natural language input element is configured to provide the received inputs
However, Serdiukova discloses
wherein the user interface comprises a natural language input element for receiving corrective inputs in natural language format, wherein the natural language input element is configured to provide the received inputs (Fig 5 “other” and [0039] “the user may provide the AI with feedback on the subject lines, for example, by interacting with a thumbs up icon, a thumbs down icon, and/or by providing unstructured feedback in a text prompt. Other forms and mechanisms for providing feedback are also contemplated within the scope of the present disclosure”)
Serdiukova suggests it is advantageous to include wherein the user interface comprises a natural language input element for receiving corrective inputs in natural language format, wherein the natural language input element is configured to provide the received inputs, because doing so can provide a convenient and intuitive mechanism to help ensure the generated content is desirable to the liking of the content creator ([0018] & [0029] & [0039] & [0053]-[0054]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Lohiya in view of Naveh to include wherein the user interface comprises a natural language input element for receiving corrective inputs in natural language format, wherein the natural language input element is configured to provide the received inputs, as taught by Serdiukova, because doing so can provide a convenient and intuitive mechanism to help ensure the generated content is desirable to the liking of the content creator.
With respect to claim 21, Lohiya and Naveh teach the method of claim 1. Lohiya does not appear to disclose,
processing the new content item by a policy check system to evaluate the new content item for sensitive material, wherein the policy check system screens out new content items that violate a platform policy
However, Serdiukova discloses
processing the new content item by a policy check system to evaluate the new content item for sensitive material, wherein the policy check system screens out new content items that violate a platform policy ([0050] “may employ a toxicity evaluator to ensure that the candidate subject lines returned at 415 include safe results. For example, instead of checking a candidate subject line against a list of inappropriate or toxic words and removing candidate subject lines with words or phrases that could be construed as offensive, the subject line generator 440 may actively create, filter, and rank candidate subject lines on the basis of toxicity (i.e., the cumulative or combined toxicity of words or phrases in a given subject line) to reduce the overall toxicity of candidate subject lines generated by the subject line generator 440”, [0061] “The content generation service 125-c may also scan the reference string to determine whether the reference string includes any words or phrases that could be construed as offensive or inappropriate. If any offensive terms are identified, the content generation service 125-c may send an alert message to the cloud client” – Examiner notes the phrase “wherein the policy check system screens out new content items that violate a platform policy” merits no patentable weight, as this fails to set forth a method step. It is descriptive material, and it does not necessarily describe how the evaluation function is performed – In an effort to expedite prosecution, Examiner notes that prior art reference Omar (cited at the end of this action, and not relied upon) discloses this feature as well)
Serdiukova suggests it is advantageous to include processing the new content item by a policy check system to evaluate the new content item for sensitive material, wherein the policy check system screens out new content items that violate a platform policy, because doing so can ensure the content is not offensive, which can increase user satisfaction ([0050], [0061]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Lohiya in view of Naveh to include processing the new content item by a policy check system to evaluate the new content item for sensitive material, wherein the policy check system screens out new content items that violate a platform policy, as taught by Serdiukova, because doing so can ensure the content is not offensive, which can increase user satisfaction.
Prior Art of Record
The prior art made of record and not relied upon is considered pertinent to the applicant’s disclosure.
Tayeb et al. (U.S. PG Pub No. 2025/0363525, November 27, 2025) teaches a system that generates personalized marketing content for target audience groups using generative models.
Mendelson et al. (U.S. PG Pub No. 2025/0363521, November 27, 2025) teaches a system that generates personalized marketing content for target audience groups using generative models.
West et al. (U.S. PG Pub No. 2024/0320710, September 26, 2024) teaches a system that generates personalized marketing content for target audience groups using generative models.
McDevitt (U.S. PG Pub No. 2024/0296482, September 5, 2024) teaches a system that generates personalized marketing content for specific recipients using generative models.
Omar (U.S. Patent No. 12,105,755) teaches processing the new content item by a policy check system to evaluate the new content item for sensitive material, wherein the policy check system screens out new content items that violate a platform policy
Conclusion
No claim is allowed
THIS ACTION IS MADE FINAL. 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 JAMES M DETWEILER whose telephone number is (571)272-4704. The examiner can normally be reached on Monday-Friday from 8 AM to 5 PM ET.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Waseem Ashraf can be reached at telephone number (571)-270-3948. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from Patent Center. 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).
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) Form at https://www.uspto.gov/patents/uspto-automated- interview-request-air-form.
/JAMES M DETWEILER/Primary Examiner, Art Unit 3621
Prosecution Insights