Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
DETAILED ACTION
The Action is responsive to the Amendments and Remarks filed on 5/7/2026. Claims 1-30 are pending claims. Claims 1, 12, and 23 are written in independent form. Claims 24-30 are newly added.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-5, 7-16, and 18-23 are rejected under 35 U.S.C. 103 as being unpatentable over Chrysanthou (U.S. Pre-Grant Publication No. 2024/0176839) and further in view of International Publication Number WO 2025/042852A1, hereinafter referred to as Barraclough and Ciminelli et al. (U.S. Pre-Grant Publication No. 2023/0409298, hereinafter referred to as Ciminelli).
Regarding Claim 1:
Chrysanthou teaches a computer-implemented method comprising:
Receiving a user-generated query comprising a natural language prompt;
Chrysanthou teaches “the user inputs a description using their user computing device 104, as described herein” (Para. [0112]) where the user input description is “of a business or organization” where the user input description is a query/request to dynamically generate a website (Paras. [0042]-[0043]).Chrysanthou further teaches “At step 300, the user inputs a description using their user computing device 104, as described herein. At step 302, the NLP engine 206 processes the description text string to extract key terms, as described herein.” (Para. [0112]) thereby teaching the user-generated query comprising a natural language prompt requiring natural language processing by NLP engine 206.
Determining an intent of the query,
Chrysanthou teaches determining “suggested key term(s)” related to the user input description (Para. [0113]) thereby teaching determining an intent of the query.
Modifying the query based on the determined intent to result in a contextualized query,
Chrysanthou teaches creating a contextualized query in the form of an “approved key term list” by modifying the user inputs description 300 to only include key terms (Fig. 3 & Paras. [0112]-[0114])
the contextualized query specifying (i) one or more data sources to be searched and (ii) one or more content types to be obtained;
Chrysanthou teaches “the Internet scanning engine 208 ranks each potentially relevant third-party website using at least one of a matching algorithm and a traffic data algorithm, so that a weighted analysis can be performed.” (Para. [0068]) where “for example, if a threshold percentage of terms in the approved key term list appear on the third-party website or its source code, then it may be deemed to be a potentially relevant website.” (Para. [0069]) and “for example, if the approved key term list includes the terms “coffee shop” and “Chicago”, the Internet scanning engine 208 searches the Internet for websites related to coffee shops in Chicago” (Para.[0064]) thereby teaching using the query key term list to specify which data sources to be searched and the content types to be obtained.
Performing, using the contextualized query, an Internet search and receiving content responsive to the contextualized query,
Crysanthou teaches “scanning the Internet using an approved key term list” (Fig. 4 & Para. [0115]) thereby teaching Performing an Internet search using the contextualized query. Crysanthou further teaches “the Internet scanning engine 208 performs keyword-based searching against text, copy, content, headers, descriptors, and tags of third-party websites to identify websites relevant to the user's business or organization” (Para. [0115]) thereby teaching receiving content responsive to the key term list (contextualized query).
Crysanthou further teaches “the content generation engine 214 searches the Internet to identify multimedia content and text content which is similar to, but not identical to, the multimedia content and text content found on the relevant third-party websites identified by the Internet scanning engine 208, as described herein.” (Para. [0123]).
dynamically generating a new webpage responsive to the query, and
Chrysanthou teaches “the template generation engine 210 can generate a website template based on the aggregate weighted template information from the relevant third-party websites” (Para. [0086]) and “The template generation engine 210 can select a pre-defined template from the template database 228 which most closely matches the aggregate weighted template information. In this embodiment, the template generation engine 210 can edit or modify the template to better match the aggregate weighted template information more closely” (Para. [0087]).Chrysanthou further teaches “the template generation engine 210 can generate multiple website templates, and provide each of these in a side-by-side on the user computing device 104 for the user to review and select.” (Para. [0094]) and “The modified pre-defined template is then stored in the template database 228 by the template generation engine 210. In an embodiment, the template database 228 is continually updated by the template generation engine 210, such that the pre-defined templates are added to and/or modified over time, and analyzed by the artificial intelligence engine 202.” (Para. [0087]) there
merging results from the Internet search and the generating a new webpage to generate a search result page; and
Chrysanthou teaches “the template generation engine 210 can generate multiple website templates, and provide each of these in a side-by-side on the user computing device 104 for the user to review and select” (Para. [0094]) thereby teaching merging the combined results from the template generation engine from the third-part websites and the pre-defined template database to generate a search results page.
Crysanthou further teaches the internet search for existing webpages as part of merging results from different services by teaching “the Internet scanning engine 208 performs keyword-based searching against text, copy, content, headers, descriptors, and tags of third-party websites to identify websites relevant to the user's business or organization. In addition, the Internet scanning engine 208 searches source code, such as HTML code and metadata, of third-party websites to identify relevant websites matching the approved key term list, or including key terms in the approved key term list, as described herein.” (Para. [0115]) and “the third-party services 106 are communicatively coupled to the server 100 via a network. The server 100 may utilize an API which allows the server 100 to communicate with each third-party service 106. Each third-party service 106 may require the server 100 to utilize a different API in order to communicate with, and receive data from, that respective the third-party service 106.” (Para. [0039]).
Dynamically generating, derived from the received content responsive to the contextualized query, at least one webpage responsive to the user-generated query,
Chrysanthou teaches “the content generation engine 214 creates a website using the identified multimedia content, text content, and/or social media content” (Para. [0125]) where at least the multimedia content and text content are derived based on a being relevant to “content found on the relevant third-part websites identified by the Internet scanning engine” (Paras. [0121]-[0123]).
The LLM generating a page outline defining content sections and a writing plan, and
Chrysanthou teaches “a template generation engine 210 analyzes each relevant third-party website for various template information” (Para. [0076]) where “The structure and layout of elements on the relevant third-party website, such as, for example, the location of the fold, use of frames, header and footer locations, menu and menu bar locations, social media link locations, widget locations, and the like.” (Para. [0083]) and “The template generation engine 210 creates a website template taking into account at least one of: the types of pages contained on the identified websites, the types of content included on the identified websites, the context of the content on the identified websites, what the content on the identified websites depict, the color schemes of the identified websites, the font and font sizes utilized on the identified websites, the structure and layout of elements on the identified websites, and the social media information utilized on the identified websites.” (Para. [0086]).Therefore, Chyrsanthou teaches generating a page outlining content sections and a writing plan via structure and layout of elements, types of content, context, what the content depicts, font and font size, etc.
generating content on a section-by-section basis according to the writing plan.
Chrysanthou teaches “the approved website template is processed by a content generation engine 214 to create a website. In an embodiment, the content generation engine 214 can generate various content for the website, such as, for example, images, videos, text, and copy.”
Crysanthou explicitly teaches all of the elements of the claimed invention as recited above except:
Determining, using a large language model (LLM), an intent of the query;
The prompt being classified by the LLM into one of a plurality of intent categories comprising seeking information, seeking products, seeking a website, or seeking images;
Modifying, by the LLM, the query;
Performing user interface generation steps in parallel.
Dynamically generating, by the LLM, at least one webpage;
However, in the related field of endeavor of automated website content generation, Barraclough in combination with Chrysanthou teaches:
Determining, using a large language model (LLM), an intent of the query;
Barraclough teaches “Natural Language Processing (NLP) techniques are employed to understand the context and intent behind user inputs and retrieved information.” (Para. [00168]) where LLMs are known NLP techniques.
The prompt being classified by the LLM into one of a plurality of intent categories comprising seeking information, seeking products, seeking a website, or seeking images;
Barraclough teaches “techniques are employed to understand the context and intent behind user inputs and retrieved information.” (Para. [00169]) and “The logic sends user context to pre-trained large language models (LLMs) such as OpenAI's GPT-3.5 or GPT4 111, Anthropic's Claude 109, Azure Al 110 and Google Vertex Al 112 for generating text, code and assets. One or more such LLMs can be used for different purposes and/or disparately managed (e.g., different ones may produce better results in terms of generating certain types of texts, video, and/or graphics) and in some cases the same purposes with different results useful for the logic 114 to assess and compare to the user’s input and media-based assets” (Para. [0047]).Barraclough further teaches “the collected data is injected into prompts for Large Language Models to steer them into producing website content, page layouts, and component configurations matching the user's identity and business niche. For example, a wedding photography business has different site needs compared to a sports photography brand. The onboarding context focuses the generated output to align with the user's specific goals.” (Para. [0067]) and “The existing context is sufficient for the Large Language Models to generate text, components, and layouts tailored to the user's needs for the specified page type” (Para. [0068]).
Modifying, by the LLM, the query;
Barraclough teaches “The most relevant retrieved information is used to enhance the prompts sent to the large language models (LLMs) described in Example Embodiment 4. b. This creates more context-rich, tailored prompts that guide the LLMs to generate highly relevant content.” (Para. [00168]) thereby teaching as part of the LLM system, modifying the query to enhance prompts.
Dynamically generating, by the LLM, at least one webpage;
Barraclough teaches “dynamically selecting and invoking Large Language Models (LLMs) to generate content” for webpages (Para. [0059]).
Thus, it would have been obvious to one of ordinary skill in the art, having the teachings of Barraclough and Chrysanthou at the time that the claimed invention was effectively filed, to have combined the use of one or more LLMs in generating a website, as taught by Barraclough, with the systems and methods for automatically generating websites using artificial intelligence, as taught by Crysanthou.
One would have been motivated to make such combination because Barraclough teaches the benefits of an LLM as “An LLM is trained on large amounts of data (e.g., millions of gigabytes of text and/or metadata) from the Internet or from other large data sources, to recognize and generate phrases, interpret languages, discern contexts being used in communications, and understand how words, sentences, intonations and characters work together.” (Para. [0045]) where “one or more such LLMs can be used for different purposes and/or disparately managed (e.g., different ones may produce better results in terms of generating certain types of texts, video, and/or graphics) and in some cases the same purposes with different results useful for the logic 114 to assess and compare to the user’s input and media-based assets and, in certain circumstances, provide as feedback to the user.” (Para. [0047]). It would have been obvious to a person having ordinary skill in the art that using different LLMs to provide different results would create a more versatile system and increase the likelihood of producing user-satisfactory results when generating the website.It is further noted that Crysanthou explicitly recites “using artificial intelligence” without providing further clarity on the types of AI systems being used and Barraclough teaches LLMs as a type of AI program: “As noted with the above discussion, and in connection with further more specific examples described below and in connection with the figures, such aspects include reference to Al and/or ML with Al and ML referring to artificial intelligence and machine learning, respectively, and appreciating that Al may be used interchangeably with ML, and to LLMs (Large language models) which are Al programs that use deep learning to analyze and understand text” (Para. [0046]).
Barraclough and Crysanthou explicitly teach all of the elements of the claimed invention as recited above except:
Performing user interface generation steps in parallel.
However, in the related field of endeavor of generating user interfaces from free text, Ciminelli in combination with Barraclough and Crysanthou teaches:
Performing user interface generation steps in parallel.
Ciminelli teaches “In embodiments of the presently disclosed subject matter, one or more stages illustrated in the figures may be executed in a different order and/or one or more groups of stages may be executed simultaneously.” (Para. [0032])
Thus, it would have been obvious to one of ordinary skill in the art, having the teachings of Ciminelli, Barraclough, and Chrysanthou at the time that the claimed invention was effectively filed, to have combined the use of target audience preferences/susceptibilities, as taught by Ciminelli, with the use of one or more LLMs in generating a website, as taught by Barraclough, and the systems and methods for automatically generating websites using artificial intelligence, as taught by Crysanthou.
One would have been motivated to make such combination because while Chrysanthou teaches
Scanning third-party websites for various template information, including “color scheme and/or palette of the relevant third-party website, such as, for example, the background color(s), header color(s), font color(s), hyperlink and mouseover color(s), and the like. f. The font(s) and font size(s) used on the relevant third-party website, such as, for example, various heading fonts, copy fonts, footer fonts, hyperlink fonts, and the like.” (Paras. [0081]-[0082]), Ciminelli teaches using known characteristics of the target audience such as “susceptibility to styles and/or designs and/or layouts and/or design elements and/or textual messages and/or color schemes and/or fonts, demographic characteristics (such as age group, gender, ethnicity, religion, income level, education level, etc.) associated with the target audience, inclination to taking specific actions, behavior patterns, languages, and so forth.” (Para. [0038]) and it would have been obvious to a person having ordinary skill in the art that utilizing known preferences of a target audience to personalize the generated websites to the target audience would increase the approval by the audience when viewing the generated websites.
Regarding Claim 2:
Ciminelli, Barraclough, and Chrysanthou further teach:
Wherein there are a plurality of different webpages generated which are responsive to the user-generated query.
Chrysanthou teaches “the content generation engine 214 can generate multiple websites, and provide each of these websites in a side-by-side fashion on the user computing device 104 for the user to review and select.” (Para. [0110]).
Barraclough further teaches “the system prompts users for initial website goals, topics, target audience and branding preferences. The system then automatically determines an optimal set and sequence of pages and sections to build based on the website goals, wherein for each section, the system selects appropriate UI elements, components and styles” (Para. [0043]).
Regarding Claim 3:
Ciminelli, Barraclough, and Chrysanthou further teach:
Wherein each different webpage is generated by the LLM using a different page generation strategy.
Barraclough teaches “the system prompts users for initial website goals, topics, target audience and branding preferences. The system then automatically determines an optimal set and sequence of pages and sections to build based on the website goals, wherein for each section, the system selects appropriate UI elements, components and styles” (Para. [0043]) thereby teaching generating a plurality of different webpages responsive to the query/user input.Barraclough further teaches “the existing context is sufficient for the Large Language Models to generate text, components, and layouts tailored to the user's needs for the specified page type” (Para. [0068]) thereby teaching using different page generation strategies to tailor layouts based at least in part on a specified page type of the “optimal set and sequence of pages and sections to build” (Para. [0043]).
Regarding Claim 4:
Ciminelli, Barraclough, and Chrysanthou further teach:
Inputting the contextualized query into the LLM and obtaining each of the different page generation strategies;
Barraclough teaches “the system dynamically selects an appropriate set of content blocks to populate each page section. The blocks are chosen from the component library based on properties like layout format, text length, media types to match the content needs for that specific page section.” (Para. [00106])
Wherein at least a portion of the dynamically generated at least one webpage comprises sections derived from different page generation strategies.
Barraclough teaches “the page, titled "Blush and Brilliance," features a cohesive and visually appealing layout that highlights the bridesmaids' role in the wedding celebration. As indicated in such a screenshot, the Al-driven system can intelligently curate a collection of wedding photos, arranging them in an elegant grid layout that complements the overall design” (Para. [00190]), thereby teaching multiple sections derived from different page generation strategies including layout.
Barraclough also teaches “the system dynamically selects an appropriate set of content blocks to populate each page section. The blocks are chosen from the component library based on properties like layout format, text length, media types to match the content needs for that specific page section.” (Para. [00106]) thereby teaching multiple content blocks, each chosen based on at least a layout property, where “each page section” can comprise multiple content blocks.
Regarding Claim 5:
Ciminelli, Barraclough, and Chrysanthou further teach:
Wherein the page generation strategies specify content types and layout for the corresponding webpage.
Barraclough teaches “the system dynamically selects an appropriate set of content blocks to populate each page section. The blocks are chosen from the component library based on properties like layout format, text length, media types to match the content needs for that specific page section.” (Para. [00106])
Regarding Claim 7:
Ciminelli, Barraclough, and Chrysanthou further teach:
Wherein the at least one webpage comprises different sections generated by the LLM using different page generation strategies.
Barraclough teaches “the page, titled "Blush and Brilliance," features a cohesive and visually appealing layout that highlights the bridesmaids' role in the wedding celebration. As indicated in such a screenshot, the Al-driven system can intelligently curate a collection of wedding photos, arranging them in an elegant grid layout that complements the overall design” (Para. [00190]), thereby teaching multiple sections derived from different page generation strategies including layout.
Barraclough also teaches “the system dynamically selects an appropriate set of content blocks to populate each page section. The blocks are chosen from the component library based on properties like layout format, text length, media types to match the content needs for that specific page section.” (Para. [00106]) thereby teaching multiple content blocks, each chosen based on at least a layout property, where “each page section” can comprise multiple content blocks.
Regarding Claim 8:
Ciminelli, Barraclough, and Chrysanthou further teach:
Inputting the received content responsive to the contextualized query into the LLM and receiving an output of the LLM;
Barraclough teaches “The most relevant retrieved information is used to enhance the prompts sent to the large language models (LLMs) described in Example Embodiment 4. b. This creates more context-rich, tailored prompts that guide the LLMs to generate highly relevant content” and “the LLM generates initial content based on the enhanced prompts” (Para. [00168])
Wherein the dynamically generated at least one webpage is derived from the output of the LLM.
Barraclough teaches “Content Generation and Augmentation: a. The LLM generates initial content based on the enhanced prompts. b. The system then augments this content by intelligently incorporating key information from the retrieved data.” (Para. [00168])
Regarding Claim 9:
Ciminelli, Barraclough, and Chrysanthou further teach:
Searching pre-existing webpages for content responsive to the contextualized query and obtaining the content responsive to the contextualized query;
Chrysanthou teaches “the Internet scanning engine 208 performs keyword-based searching against text, copy, content, headers, descriptors, and tags of third-party websites to identify websites relevant to the user's business or organization” (Para. [0115]) thereby teaching searching pre-existing third-party websites for content relevant to the contextualized key term list/query, and obtaining content responsive to the searching.
Wherein the dynamically generated at least one webpage is derived from at least one pre-existing webpage having matching content responsive to the contextualized query.
Chrysanthou teaches “the template generation engine 210 analyzes each relevant third-party website for various template information, such as, but not limited to, the types of pages, the type of content on each page, the context of the content, what the content depicts, the color scheme and/or palette, the font(s) and font size(s) used, the structure and layout of elements, and/or the presence or absence of social media content, as described herein.” (Para. [0117]), “generates a website template based on the template information determined the relevant third-party websites,” (Para. [0118]), and “generating a website using the website template” (Para. [0121]).
Regarding Claim 10:
Ciminelli, Barraclough, and Chrysanthou further teach:
Determining, by the LLM, follow up questions to content in the at least one webpage;
Chrysanthou teaches follow up questions after generating a website by teaching “the publishing engine 210 can prompt a number of questions, such as, for example, “Do you like the color scheme?”, “Do you like the font?”, “Do you like the layout?”, and the like, and the pre-defined answers can be, for example, “Yes”, “No”, “Somewhat”, and the like. In another embodiment, the user can be prompted to provide feedback based on a scale of one to ten for various aspects of the website.” (Para. [0105]).
Generating, by the LLM, additional content based on the determined follow up questions; and
Chrysanthou teaches “based on the user feedback, the template generation engine 210 and/or the content generation engine 214 can generate another website template or another website by considering the user feedback.” (Para. [0106]). Where
Enriching the at least one webpage with at least a portion of the generated additional content.
Chrysanthou teaches “based on the user feedback, the template generation engine 210 and/or the content generation engine 214 can generate another website template or another website by considering the user feedback.” (Para. [0106]).
Regarding Claim 11:
Ciminelli, Barraclough, and Chrysanthou further teach:
Determining, by the LLM, that an Internet search for content responsive to the follow up questions is required;
Barraclough teaches “The user can iteratively improve the site through feedback prompts which trigger new generation cycles” (Para. [0047]) and “allows users to iteratively modify and refine generated website content through interactive feedback. For an existing page built using automated Al generation or manual user creation, the system presents options to edit styles, text, media elements, and to rebuild blocks” (Para. [00128]) thereby determining that a new generation cycle, and thus an internet search for content responsive to follow up feedback questions, is required.
Generating, by the LLM, one or more follow up question queries;
Barraclough teaches “allows users to iteratively modify and refine generated website content through interactive feedback. For an existing page built using automated Al generation or manual user creation, the system presents options to edit styles, text, media elements, and to rebuild blocks” (Para. [00128]) thereby teaching generating follow up question queries to provide content when rebuilding blocks for the next iteration of generation cycles.
Performing a second Internet search and receiving content responsive to the one or more follow up question queries; and
Chrysanthou teaches “the content generation engine 214 searches the Internet to identify multimedia content and text content” (Para. [0123]) as part of the website generation cycle. Therefore, Barraclough in combination with Chrysanthou teaches performing a second internet search, based on an interactive feedback system, and receiving content when rebuilding blocks for the next iteration of generation cycles.
Enriching the at least one webpage with content generated by the LLM based on the second Internet search.
Chrysanthou teaches “based on the user feedback, the template generation engine 210 and/or the content generation engine 214 can generate another website template or another website by considering the user feedback.” (Para. [0106]) thereby teaching enriching the webpage by considering the user feedback in the next iteration of generation cycles taught by Barraclough (Paras. [0047] & [00128]), the next iteration of generation cycles including the internet search taught by Chrysanthou (Para. [0123]).
Regarding Claim 12:
Some of the limitations herein are similar to some or all of the limitations of Claim 1.
Ciminelli, Barraclough, and Chrysanthou further teach a system comprising:
At least one data processor (Chrysanthou – Para. [0007]); and
Memory storing instructions which, when executed by the at least one data processor, result in operations (Chrysanthou – Para. [0007]).
Regarding Claim 13:
All of the limitations herein are similar to some or all of the limitations of Claim 2.
Regarding Claim 14:
All of the limitations herein are similar to some or all of the limitations of Claim 3.
Regarding Claim 15:
All of the limitations herein are similar to some or all of the limitations of Claim 4.
Regarding Claim 16:
All of the limitations herein are similar to some or all of the limitations of Claim 5.
Regarding Claim 18:
All of the limitations herein are similar to some or all of the limitations of Claim 7.
Regarding Claim 19:
All of the limitations herein are similar to some or all of the limitations of Claim 8.
Regarding Claim 20:
All of the limitations herein are similar to some or all of the limitations of Claim 9.
Regarding Claim 21:
All of the limitations herein are similar to some or all of the limitations of Claim 10.
Regarding Claim 22:
All of the limitations herein are similar to some or all of the limitations of Claim 11.
Regarding Claim 23:
Chrysanthou teaches a computer-implemented method comprising:
Receiving, over a network from a remote computing device, a user-generated query;
Chrysanthou teaches “the user inputs a description using their user computing device 104, as described herein” (Para. [0112]) where the user input description is “of a business or organization” where the user input description is a query/request to dynamically generate a website (Paras. [0042]-[0043]).Chrysanthou further teaches “The network architecture includes a server 100, a website generator platform 102 coupled to the server 100 via a network connection, at least one user computing device 104 coupled to the server 100 via a network connection, and third-party services 106 (individually, services 106-1 and 106-2) coupled to the server 100 via a network connection.” (Para.[0027]) and “the intake engine 204 displays a text input field on the user computing device 104 that allows the user to input the description” (Para. [0044]).
Determining an intent of the query;
Chrysanthou teaches determining “suggested key term(s)” related to the user input description (Para. [0113]).
Dynamically generating, derived from the received content responsive to the corresponding contextualized query, at least one webpage.
Chrysanthou teaches “the content generation engine 214 creates a website using the identified multimedia content, text content, and/or social media content” (Para. [0125]) where at least the multimedia content and text content are derived based on a being relevant to “content found on the relevant third-part websites identified by the Internet scanning engine” (Paras. [0121]-[0123]).
Crysanthou explicitly teaches all of the elements of the claimed invention as recited above except:
Determining, using a large language model (LLM) being executed on a server, an intent of the query;
Modifying, by the LLM, the query based on the determined intent to result in a plurality of contextualized queries, each of the contextualized queries corresponding to a different webpage generation strategy,
Each webpage generation strategy comprising instructions to generate a webpage including how content is obtained over the Internet,
Each webpage generation strategy comprising specifications for a user interface for conveying information,
At least two of the webpage generation strategies specifying different workflows for obtaining content;
Performing, for each webpage generation strategy, an Internet search according to the corresponding workflow for the webpage generation strategy and receiving content responsive to the corresponding contextualized query; and
Dynamically generating, by the LLM and derived from the received content responsive to the corresponding contextualized query, at least one webpage for each webpage generation strategy responsive to the user-generated query and causing at least one webpage corresponding to two or more webpage generation strategies to be viewed on the remote computing device.
However, in the related field of endeavor of automated website content generation, Barraclough in combination with Chrysanthou teaches:
Determining, using a large language model (LLM) being executed on a server, an intent of the query;
Chrysanthou teaches determining “suggested key term(s)” related to the user input description (Para. [0113]) and Barraclough teaches “Natural Language Processing (NLP) techniques are employed to understand the context and intent behind user inputs and retrieved information.” (Para. [00168]) where LLMs are known NLP techniques.Barraclough further teaches “it will be apparent that various known devices may be used with the aspects and features described herein for example embodiments. As non-limiting examples, such devices may include one or more in combination of the following: devices including communications circuits such as servers, user-operable (e.g., network-enabled) devices such as computer processing circuits (e.g., smart phones and other personal assistant devices (aka user endpoint devices) with user interfaces, laptops, desk-based computer etc.).” (Para. [00163]).
Therefore, Barraclough in combination with Chrysanthou teaches determining, using an LLM being executed on a server, an intent of the query.
Modifying, by the LLM, the query based on the determined intent to result in a plurality of contextualized queries, each of the contextualized queries corresponding to a different webpage generation strategy,
Chrysanthou teaches creating a contextualized query in the form of an “approved key term list” by modifying the user inputs description 300 to only include key terms (Fig. 3 & Paras. [0112]-[0114]) and Barraclough teaches “The most relevant retrieved information is used to enhance the prompts sent to the large language models (LLMs) described in Example Embodiment 4. b. This creates more context-rich, tailored prompts that guide the LLMs to generate highly relevant content.” (Para. [00168]). Barraclough further teaches “the LLM Response (402) is structured to receive and organize the output generated by the LLM. It consists of multiple LLM Block Property Value sections (406), each corresponding to a specific prompt in the User Prompt section.” (Para. [0057]).
Therefore, Barraclough in combination with Chrysanthou teaches modifying, by the LLM system, the query to enhance prompts based on different LLMs/webpage generation strategies corresponding to different prompts in the user prompt section and the determined intent to result in a contextualized query for each of the different LLMs.
Barraclough further teaches “variants of the website can be built and maintained for localized markets or alternate audiences. This embodiment enables automatically adapting an existing website to new languages and geographies by leveraging Al translation, localization and content generation capabilities. Variants can also be generated for different demographic groups/audiences that are not necessarily defined by geography, but by other attributes where a site that provides targeted content and SEO would be advantageous” (Para. [00154]) thereby teaching different webpage generation strategies for specifying different workflows for obtaining content suited for the different demographic groups/audiences to provide targeted content.
Each webpage generation strategy comprising instructions to generate a webpage including how content is obtained over the Internet,
Barraclough teaches “The most relevant retrieved information is used to enhance the prompts sent to the large language models (LLMs) described in Example Embodiment 4. b. This creates more context-rich, tailored prompts that guide the LLMs to generate highly relevant content.” (Para. [00168]).
At least two of the webpage generation strategies specifying different workflows for obtaining content;
Barraclough teaches the webpage generation strategies specifies different workflows for obtaining content by teaching “The LLM Invoker is the central component that interfaces with various LLM providers through their respective Software Development Kits (SDKs) as depicted in the (506 through N). Each SDK connects to its corresponding LLM service, such as Open Al , Google Vertex, Claude, and Azure (e.g., as in FIG. 5). A Self-hosted LLM option (507) is used to locally run custom fine-tuned models within the infrastructure instead of using cloud providers.” (Para. [0061]).Barraclough further teaches “variants of the website can be built and maintained for localized markets or alternate audiences. This embodiment enables automatically adapting an existing website to new languages and geographies by leveraging Al translation, localization and content generation capabilities. Variants can also be generated for different demographic groups/audiences that are not necessarily defined by geography, but by other attributes where a site that provides targeted content and SEO would be advantageous” (Para. [00154]) thereby teaching different webpage generation strategies for specifying different workflows for obtaining content suited for the different demographic groups/audiences to provide targeted content via different website variants.
Performing, for each webpage generation strategy, an Internet search according to the corresponding workflow for the webpage generation strategy and receiving content responsive to the corresponding contextualized query; and
Crysanthou teaches “scanning the Internet using an approved key term list” (Fig. 4 & Para. [0115]) thereby teaching Performing an Internet search using the contextualized query. Crysanthou further teaches “the Internet scanning engine 208 performs keyword-based searching against text, copy, content, headers, descriptors, and tags of third-party websites to identify websites relevant to the user's business or organization” (Para. [0115]) thereby teaching receiving content responsive to the key term list (contextualized query). Crysanthou further teaches “the content generation engine 214 searches the Internet to identify multimedia content and text content which is similar to, but not identical to, the multimedia content and text content found on the relevant third-party websites identified by the Internet scanning engine 208, as described herein.” (Para. [0123]).
Barraclough further teaches “variants of the website can be built for different demographic groups/audiences that are not necessarily defined by geography, but by other attributes where a site that provides targeted content and SEO would be advantageous” (Para. [00154]) and “The most relevant retrieved information is used to enhance the prompts sent to the large language models (LLMs) described in Example Embodiment 4. b. This creates more context-rich, tailored prompts that guide the LLMs to generate highly relevant content.” (Para. [00168]) thereby teaching different webpage generation strategies for specifying different workflows for obtaining content suited for the different demographic groups/audiences to provide targeted content via different website variants.
Dynamically generating, by the LLM and derived from the received content responsive to the corresponding contextualized query, at least one webpage for each webpage generation strategy responsive to the user-generated query and causing at least one webpage corresponding to two or more webpage generation strategies to be viewed on the remote computing device.
Chrysanthou teaches “the content generation engine 214 creates a website using the identified multimedia content, text content, and/or social media content” (Para. [0125]) where at least the multimedia content and text content are derived based on a being relevant to “content found on the relevant third-part websites identified by the Internet scanning engine” (Paras. [0121]-[0123]).
Barraclough further teaches “variants of the website can be built for different demographic groups/audiences that are not necessarily defined by geography, but by other attributes where a site that provides targeted content and SEO would be advantageous” (Para. [00154]) and “The most relevant retrieved information is used to enhance the prompts sent to the large language models (LLMs) described in Example Embodiment 4. b. This creates more context-rich, tailored prompts that guide the LLMs to generate highly relevant content.” (Para. [00168]) where “the LLM Response (402) is structured to receive and organize the output generated by the LLM. It consists of multiple LLM Block Property Value sections (406), each corresponding to a specific prompt in the User Prompt section.” and “the system can produce website-specific content that can be reliably deserialized into block properties” (Paras. [0057]-[0058]) thereby teaching a plurality of webpage generation strategies in a single webpage/output corresponding in part to the specific prompt in the user prompt section for specifying different workflows for obtaining content suited for the different demographic groups/audiences to provide targeted content via different website variants.
Barraclough and Crysanthou explicitly teach all of the elements of the claimed invention as recited above except:
Each webpage generation strategy comprising specifications for a user interface for conveying information,
However, in the related field of endeavor of dynamic websites and searching content, Ciminelli teaches:
Each webpage generation strategy comprising specifications for a user interface for conveying information,
Ciminelli teaches “the information related to the target audience and/or historic activities of individuals associated with the target audience may be analyzed to determine a characteristic of the target audience, for example using a rule based analysis. Some non-limiting examples of such characteristics of the target audience may include susceptibility to styles and/or designs and/or layouts and/or design elements and/or textual messages and/or color schemes and/or fonts, demographic characteristics (such as age group, gender, ethnicity, religion, income level, education level, etc.) associated with the target audience, inclination to taking specific actions, behavior patterns, languages, and so forth.” (Para. [0038]).
Thus, it would have been obvious to one of ordinary skill in the art, having the teachings of Ciminelli, Barraclough, and Chrysanthou at the time that the claimed invention was effectively filed, to have combined the use of target audience preferences/susceptibilities, as taught by Ciminelli, with the use of one or more LLMs in generating a website, as taught by Barraclough, and the systems and methods for automatically generating websites using artificial intelligence, as taught by Crysanthou.
One would have been motivated to make such combination because while Chrysanthou teaches
Scanning third-party websites for various template information, including “color scheme and/or palette of the relevant third-party website, such as, for example, the background color(s), header color(s), font color(s), hyperlink and mouseover color(s), and the like. f. The font(s) and font size(s) used on the relevant third-party website, such as, for example, various heading fonts, copy fonts, footer fonts, hyperlink fonts, and the like.” (Paras. [0081]-[0082]), Ciminelli teaches using known characteristics of the target audience such as “susceptibility to styles and/or designs and/or layouts and/or design elements and/or textual messages and/or color schemes and/or fonts, demographic characteristics (such as age group, gender, ethnicity, religion, income level, education level, etc.) associated with the target audience, inclination to taking specific actions, behavior patterns, languages, and so forth.” (Para. [0038]) and it would have been obvious to a person having ordinary skill in the art that utilizing known preferences of a target audience to personalize the generated websites to the target audience would increase the approval by the audience when viewing the generated websites.
Regarding Claim 24:
Ciminelli, Barraclough, and Chrysanthou further teach:
Wherein each webpage generation strategy comprises a writing plan specifying content types and content sources for the corresponding webpage,
Chrysanthou teaches “a template generation engine 210 analyzes each relevant third-party website for various template information” (Para. [0076]) where “The structure and layout of elements on the relevant third-party website, such as, for example, the location of the fold, use of frames, header and footer locations, menu and menu bar locations, social media link locations, widget locations, and the like.” (Para. [0083]) and “The template generation engine 210 creates a website template taking into account at least one of: the types of pages contained on the identified websites, the types of content included on the identified websites, the context of the content on the identified websites, what the content on the identified websites depict, the color schemes of the identified websites, the font and font sizes utilized on the identified websites, the structure and layout of elements on the identified websites, and the social media information utilized on the identified websites.” (Para. [0086]).Therefore, Chyrsanthou teaches each webpage generation strategy for each “relevant third-party website” including a writing plan that specifies structure and layout of elements, types of content, context, what the content depicts, font and font size, etc.
Wherein the contextualized query specifies which data sources from which to obtain content and content types to be obtained.
Chrysanthou teaches “the Internet scanning engine 208 ranks each potentially relevant third-party website using at least one of a matching algorithm and a traffic data algorithm, so that a weighted analysis can be performed.” (Para. [0068]) where “for example, if a threshold percentage of terms in the approved key term list appear on the third-party website or its source code, then it may be deemed to be a potentially relevant website.” (Para. [0069]) and “for example, if the approved key term list includes the terms “coffee shop” and “Chicago”, the Internet scanning engine 208 searches the Internet for websites related to coffee shops in Chicago” (Para.[0064]) thereby teaching using the query key term list to specify which data sources to be searched and the content types to be obtained.
Regarding Claim 25:
Ciminelli, Barraclough, and Chrysanthou further teach:
Wherein at least one of the webpage generation strategies comprises a deep-dive information strategy in which the LLM generates a webpage outline defining a layout of the webpage including content sections and a writing plan specifying content types and content sources, and
Chrysanthou teaches “a template generation engine 210 analyzes each relevant third-party website for various template information” (Para. [0076]) where “The structure and layout of elements on the relevant third-party website, such as, for example, the location of the fold, use of frames, header and footer locations, menu and menu bar locations, social media link locations, widget locations, and the like.” (Para. [0083]) and “The template generation engine 210 creates a website template taking into account at least one of: the types of pages contained on the identified websites, the types of content included on the identified websites, the context of the content on the identified websites, what the content on the identified websites depict, the color schemes of the identified websites, the font and font sizes utilized on the identified websites, the structure and layout of elements on the identified websites, and the social media information utilized on the identified websites.” (Para. [0086]).Therefore, Chyrsanthou teaches generating a page outlining content sections and a writing plan via structure and layout of elements, types of content, context, what the content depicts, font and font size, etc.Barraclough further teaches “generate, via an LLM (large language modeling) algorithm that is computer-executed based on the constructed context-aware prompts, website content that includes text, code, and media elements, and assemble the generated content into a website structure that is optimal to the particular user in that the website structure is tailored to goals of the particular user and is consistent with attributes of the media assets.” (Para. [0012])
generates content on a section-by-section basis according to the writing plan, and
Chrysanthou teaches “the approved website template is processed by a content generation engine 214 to create a website. In an embodiment, the content generation engine 214 can generate various content for the website, such as, for example, images, videos, text, and copy.”Barraclough further teaches “ In certain particular example implementations according to the present disclosure, the system dynamically selects an appropriate set of content blocks to populate each page section. The blocks are chosen from the component library based on properties like layout format, text length, media types to match the content needs for that specific page section” (Para. [00106]).
further generates a page title.
Barraclough teaches “In another example (also according to the present disclosure but not in the drawing), a screenshot showcases a beautifully designed preproduction web page created using the Al-driven UI modification embodiment for a wedding website. The page, titled "Blush and Brilliance," features a cohesive and visually appealing layout that highlights the bridesmaids' role in the wedding celebration” (Para. [00190])
Regarding Claim 27:
Ciminelli, Barraclough, and Chrysanthou further teach:
Wherein at least one of the web page generation strategies comprises a fresh generation strategy in which the LLM generates new content without externally sourced content.
Barraclough teaches “A Self-hosted LLM option (507) is used to locally run custom fine-tuned models within the infrastructure instead of using cloud providers.” (Para. [0061]) and “The user accesses the site editor and selects the option to construct a new page. A dialog presents page type options, including "Use Artificial Intelligence Page Creator", (b) Upon selecting the Al page builder, the user specifies the desired type of new page such as "About Us" or "Contact", and any preferences, (c) This triggers the automated page construction process. Since the user already exists in the system, their profile, website data, past page types, and media are retrieved from the servers, (d) The existing context is sufficient for the Large Language Models to generate text, components, and layouts tailored to the user's needs for the specified page type” (Para. [0068]).Barraclough further teaches generating new content by teaching “For each moment, it generates descriptive text. For example, for the "bride getting ready" moment: Prompt to LLM: "Describe the 'bride getting ready' moment of a wedding. Use these keywords from the selected images: 'excitement', 'laughter', 'sunlit room', 'white robe', 'bridesmaids'. Maintain a warm and joyful tone." Generated text: "As the morning sun streamed through the windows, the bride's suite was alive with excitement and laughter. Surrounded by her closest friends, the bride, wrapped in a soft white robe, savored these precious moments before the ceremony. The air buzzed with anticipation as the bridesmaids helped with final touches, each shared glance and smile capturing the joy of this special day."” Para. [00182]).
Regarding Claim 29:
Ciminelli, Barraclough, and Chrysanthou further teach:
Wherein the at least one webpage for each webpage generation strategy comprises an AI copilot interface enabling users to ask questions related to content on the webpage.
Barraclough teaches “The chatbot offers Large Language Model-generated page variations to select from, integrates requested features like appointment booking through Zenfolio’s BookMe service (see https://zenfolio.com/), and handles publishing and social promotion of the finalized page, enabling rapid mobile creation. In FIG. 10B, this second diagram shows editing an existing page by allowing the user to request localized changes like modifying block text style and tone or replacing media” (Para. [0077]-[0078]).Barraclough also teaches “at least one exemplary aspect allows users to iteratively modify and refine generated website content through interactive feedback. For an existing page built using automated Al generation or manual user creation, the system presents options to edit styles, text, media elements, and to rebuild blocks. These options are offered through a conversational chatbot interface. This allows the user to interact in natural language to request changes or provide feedback. Based on the user feedback intent detected via the chatbot, the system performs targeted regeneration of specific sections as needed to match the requested changes.” (Para. [00128])
Regarding Claim 30:
Ciminelli, Barraclough, and Chrysanthou further teach:
Determining, by the LLM based on content of the at least one webpage, that an Internet search for content responsive to the follow up questions is required;
Barraclough teaches “The user can iteratively improve the site through feedback prompts which trigger new generation cycles” (Para. [0047]) and “allows users to iteratively modify and refine generated website content through interactive feedback. For an existing page built using automated Al generation or manual user creation, the system presents options to edit styles, text, media elements, and to rebuild blocks” (Para. [00128]) thereby determining that a new generation cycle, and thus an internet search for content, based on the content of the webpage, responsive to follow up feedback questions, is required.
Generating, by the LLM, one or more follow up question queries based on the determined follow up questions;
Barraclough teaches “allows users to iteratively modify and refine generated website content through interactive feedback. For an existing page built using automated Al generation or manual user creation, the system presents options to edit styles, text, media elements, and to rebuild blocks” where “options are offered through a conversational chatbot interface. This allows the user to interact in natural language to request changes or provide feedback. Based on the user feedback intent detected via the chatbot, the system performs targeted regeneration of specific sections as needed to match the requested changes. (Para. [00128]) thereby teaching generating follow up question queries to regenerate specific sections based on the options offered through the conversational chatbot interface when rebuilding blocks for the next iteration of generation cycles.
Performing a second Internet search and receiving content responsive to the one or more follow up question queries; and
Chrysanthou teaches “the content generation engine 214 searches the Internet to identify multimedia content and text content” (Para. [0123]) as part of the website generation cycle. Therefore, Barraclough in combination with Chrysanthou teaches performing a second internet search, based on an interactive feedback system, and receiving content when rebuilding blocks for the next iteration of generation cycles.
Enriching the at least one webpage with content generated by the LLM based on the second Internet search.
Chrysanthou teaches “based on the user feedback, the template generation engine 210 and/or the content generation engine 214 can generate another website template or another website by considering the user feedback.” (Para. [0106]) thereby teaching enriching the webpage by considering the user feedback in the next iteration of generation cycles taught by Barraclough (Paras. [0047] & [00128]), the next iteration of generation cycles including the internet search taught by Chrysanthou (Para. [0123]).
Claim(s) 6, 17, 26, and 28 are rejected under 35 U.S.C. 103 as being unpatentable over Ciminelli, Barraclough, and Chrysanthou, and further in view of Akkiraju Venkata et al. (U.S. Pre-Grant Publication No. 2025/0005081, hereinafter referred to as Akkiraju).
Regarding Claim 6:
Ciminelli, Barraclough, and Crysanthou explicitly teach all of the elements of the claimed invention as recited above except:
Wherein the page generation strategies specify sources to search to populate content in the corresponding webpage.
However, in the related field of endeavor of dynamic websites and searching content, Akkiraju teaches:
Wherein the page generation strategies specify sources to search to populate content in the corresponding webpage.
Akkiraju teaches page generation strategies including “connector intent…to select at least one connector for connecting with corresponding at least one data store” where “the NL processor 175 generates consolidated search results including matching website content from two or more different data sources, via corresponding two or more connectors” (Para. [0021])
Thus, it would have been obvious to one of ordinary skill in the art, having the teachings of Akkiraju, Ciminelli, Barraclough, and Chrysanthou at the time that the claimed invention was effectively filed, to have combined the determination of whether a search query has connector intent for connecting with a particular data store, as taught by Akkiraju, with the use of target audience preferences/susceptibilities, as taught by Ciminelli, the use of one or more LLMs in generating a website, as taught by Barraclough, and the systems and methods for automatically generating websites using artificial intelligence, as taught by Crysanthou.
One would have been motivated to make such combination because Akkiraju teaches “universal search indexer 165 is further configured to determine, using NL processor 175, whether the search query has connector intent (i.e., a search query including an intent to select at least one connector for connecting with corresponding at least one data store). Based on a determination that the search query has connector intent, the NL processor 175 generates consolidated search results including matching website content from two or more different data sources via corresponding two or more connectors.” (Para. [0021]) and it would have been obvious to a person having ordinary skill in the art that allowing a user to provide the intent in their request/query to search a particular one or more data stores would result in improved user-satisfaction in the content results from the search.
Regarding Claim 17:
All of the limitations herein are similar to some or all of the limitations of Claim 6.
Regarding Claim 26:
Akkiraju, Ciminelli, Barraclough, and Chrysanthou further teach:
Wherein at least one of the webpage generation strategies comprises a summarization strategy in which the web is called by querying one or more search engines, and the LLM summarizes the crawled content into a structured format comprising bullet points, and
Barraclough teaches “For the text, font type, size, color scheme and alignment are programmatically determined to optimize readability, visual hierarchy and brand stylistic coherence.” and “The synthesized content, selected media asset and tailored styling align to implicitly reflect the user's brand essence and offerings.” (Para. [0074]) thereby teaching a structured format comprising tailored styling and a visually hierarchy.Akkiraju further teaches “the options field 325 includes a field for selecting whether to crawl only websites or webpages listed in the sitemap for the listed one or more URLs or to perform a full crawl of the target websites. In some cases, the options field 325 further includes a field for selecting whether to enable a crawl for dynamic websites or webpages, or whether to enable a crawl for static websites or webpages. In some instances, the options field 325 further includes a field for selecting either a crawl mode for a cloud accessible website or a crawl mode for an enterprise website or an agent of the enterprise website.” (Para. [0029]) and “The best match portion 350 includes a portion that lists one or more best-match results that may be selected for display of corresponding search results in the search results display field 355, which displays a listing of search results with links for accessing the webpages, websites, or documents associated with the query term, a summary of each search result, and/or last modified dates. (Para. [0030])
generates a page title.
Barraclough teaches “In another example (also according to the present disclosure but not in the drawing), a screenshot showcases a beautifully designed preproduction web page created using the Al-driven UI modification embodiment for a wedding website. The page, titled "Blush and Brilliance," features a cohesive and visually appealing layout that highlights the bridesmaids' role in the wedding celebration” (Para. [00190])
Regarding Claim 28:
Akkiraju, Ciminelli, Barraclough, and Chrysanthou further teach:
Wherein at least one of the webpage generation strategies comprises a crawl generation strategy in which information is obtained from the web responsive to the contextualized query, and the LLM rewrites, summarizes, or enriches the obtained information to generate webpage content distinct from the obtained information.
Barraclough teaches “For the text, font type, size, color scheme and alignment are programmatically determined to optimize readability, visual hierarchy and brand stylistic coherence.” and “The synthesized content, selected media asset and tailored styling align to implicitly reflect the user's brand essence and offerings.” (Para. [0074]) thereby teaching a structured format comprising tailored styling and a visually hierarchy. Barraclough further teaches using an LLM to rewrite or enrich the obtained information by teaching “The chatbot offers Large Language Model-generated page variations to select from, integrates requested features like appointment booking through Zenfolio’s BookMe service (see https://zenfolio.com/), and handles publishing and social promotion of the finalized page, enabling rapid mobile creation. [0078] In FIG. 10B, this second diagram shows editing an existing page by allowing the user to request localized changes like modifying block text style and tone or replacing media” (Para. [0077]-[0078]).Akkiraju further teaches “the options field 325 includes a field for selecting whether to crawl only websites or webpages listed in the sitemap for the listed one or more URLs or to perform a full crawl of the target websites. In some cases, the options field 325 further includes a field for selecting whether to enable a crawl for dynamic websites or webpages, or whether to enable a crawl for static websites or webpages. In some instances, the options field 325 further includes a field for selecting either a crawl mode for a cloud accessible website or a crawl mode for an enterprise website or an agent of the enterprise website.” (Para. [0029]) and “The best match portion 350 includes a portion that lists one or more best-match results that may be selected for display of corresponding search results in the search results display field 355, which displays a listing of search results with links for accessing the webpages, websites, or documents associated with the query term, a summary of each search result, and/or last modified dates. (Para. [0030])
Response to Amendment
Applicant’s Amendments, filed on 5/7/2026, are acknowledged and accepted.
In light of the Amendments and Remarks filed on 5/7/2026, the objections to Claims 1, 4, 8, 11, 12, 19, 22, and 23 has been withdrawn.
In light of the Amendments and Remarks filed on 5/7/2026, the 112(a) Rejection to Claims 1-23 has been withdrawn.
Response to Arguments
On page 13 of the Remarks filed on 5/7/2026, Applicant states that “Chrysanthou addresses a fundamentally different technical problem” and thus “Chrysanthou’s business website builder simply has no teaching or suggestion of generating search results in response to user queries---the entire architectural premise differs”.Applicant’s argument is not convincing because Chyrsanthou and the present application are understood to be in similar areas of invention of a website generator that utilizes artificial intelligence, as is indicated by the present specification “The subject matter described herein relates to techniques for dynamically generating and personalizing webpages leveraging advanced artificial intelligence such as large language models.” (Para. [0001]).
On pages 13-14 of the Remarks filed on 5/7/2026, Applicant argues that “Neither Chrysanthou nor Barraclough teaches the claimed parallel architecture recited in amended claims 1 and 12.” Because “Chrysanthou never generates a "new webpage" in parallel with an Internet search, nor does it merge search results with a dynamically generated page into a search result page. Barraclough similarly describes sequential website building for users, not parallel search- and-generation pipelines” and “The reliance on Ciminelli to supply "parallel" operations is misplaced; Ciminelli merely teaches that processing stages "may be executed simultaneously" in a generic sense”.Applicant’s argument is not convincing because Ciminelli does not just teach that processing stages may be executed simultaneously in a generic sense, Ciminelli is teaching that processing stages “may be executed simultaneously” in the context of processing stages that generate user interfaces from free text where Chrysanthou is understood as teaching processing stages that generate user interfaces including an internet search and generating a new webpage, as is further addressed in the rejection above.
On pages 14-15 of the Remarks filed on 5/7/2026, Applicant argues that “the prior art combination fails to teach the specific page outline and writing plan limitation recited in claims 1 and 12.” because “This is not merely generating a template-it is a two-phase LLM-driven workflow in which the LLM first generates a structured outline with a writing plan, and then executes that plan by generating content section-by-section.” and Chrysanthou teaches “fundamentally extraction and aggregation of existing content, not LLM-driven outline generation and plan-based section-by- section content creation. Barraclough's content block approach dynamically selects "content blocks" from a "component library based on properties like layout format, text length, media types" (Barraclough, paragraph [00106]), but selecting pre-built blocks from a library is categorically different from an LLM generating a writing plan and then generating content section-by-section according to that plan. Neither reference teaches the claimed outline-then- execute methodology.”.Applicant’s argument is not convincing based on the broadest reasonable interpretation of what a “writing plan” is, which is merely a plan of computer tasks/goals to be performed/achieved.
On pages 15-16 of the Remarks filed on 5/7/2026, Applicant argues that “With respect to independent claim 23, the prior art combination fails to teach the claimed multi-strategy, per-workflow retrieval architecture” because “These are not merely different LLM prompts-they represent fundamentally different workflows for obtaining content (e.g., fresh generation uses no external content, while crawl generation requires external content). The Office Action maps Barraclough's teaching that "variants of the website can be built...for different demographic groups/audiences" to the claimed different strategies. Office Action, page 26. But demographic localization variants involve adapting the same content for different audiences, not executing different content- obtaining workflows. Barraclough provides no teaching of executing materially different retrieval workflows per strategy.”Applicant’s argument is not convincing because the argument appears to be importing the scope of what “different workflows” means by providing examples from the specification “(e.g., fresh generation uses no external content, while crawl generation requires external content)” whereas the broadest reasonable interpretation must be given based on the claim language itself. It is further noted that newly added dependent claims 26-28 clarify the examples of web page generation strategies being argued, required further consideration and searching, and have been addressed in full in the rejection above.
On pages 16-17 of the Remarks filed on 5/7/2026, Applicant argues that the previously cited prior art does not teach the newly added dependent claims 24-30.The newly added dependent claims were reviewed and necessitated the new grounds of rejection presented in the rejection above.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Bradea et al. (U.S. Pre-Grant Publication No. 2025/0068893) teaches generating personalized content using generative artificial intelligence (AI) are provided. In an example method, a processing device including a personalization module receives an indication that a user interacted with content displayed on a web page. The personalization module receives a set of attributes, comprising information about the user and information about the content, and information about a segment to which the user belongs. The processing module then determines a tuning parameter, wherein the tuning parameter controls the randomness of the output of the generative AI model. The personalization module next inputs to the generative AI model the tuning parameter and a prompt comprising the set of attributes and the information about the segment and subsequently receives personalized content responsive to the tuning parameter and the prompt. The personalization module can then display the personalized content in a dynamic content field associated with the content.
Khorashadi et al. (U.S. Pre-Grant Publication No. 2014/0149850) teaches at least two web browsers operable on at least two different computing devices. A server processes requested code to return binary code as metadata to assist a computing device render a webpage. The server transmits the generated metadata to at least one computing device. The computing device renders a webpage using at least a portion of the provided metadata. The metadata may identify portions of JavaScript that can be processed in parallel. The metadata may identify a library portion that does not have to be loaded. The metadata may identify a portion of the webpage that may be rendered first before a second portion of the webpage. Returning metadata to the computing device can assist the computing device in parsing, analyzing or executing the request for the webpage.
Saxena (U.S. Pre-Grant Publication No. 2024/0330579) teaches a website development system automatically generates text for a webpage. The system obtains a prompt template associated with a section of the webpage, where the prompt template includes one or more parameters. Based on the webpage, the prompt template determines a first value for a first one of the one or more parameters. A request to provide input for a second value of a second parameter is sent for display to a user. Using the prompt template, the first value, and the second value, the system generates a prompt to a large language model to generate text for the section of the webpage.
Murali et al. (U.S. Patent No. 11,042,600) teaches techniques for enabling webpages to be generated and customized based on rules created by a user. A user accessing a user interface may select a set of webpage features, the dimensions or other characteristics of the features, and a corresponding set of rule conditions, such as characteristics of a user device, user account, connection, or webpage. When a request to access a webpage is received, an existing body of rules is queried, and the particular rule that is satisfied by the request is determined. A webpage that includes the corresponding features and feature characteristics of the satisfied rule is generated and provided to the requesting device.
Jain (U.S. Patent No. 12,072,950) teaches a dynamic unified object generation platform, including a dynamic unified object generation computer server coupled to at least a first partner and a second partner and configured to electronically receive and store in a server database at least a first set of rules from the first partner and a second set of rules from the second partner for a publisher website, a server processor coupled to the server database and configured to automatically and dynamically create a comparison between a unified set of rules and the first set of rules and the second set of rules, a pre-deployment processor coupled to the server processor and configured to dynamically convert the comparison into a single unified code structure based on the unified set of rules, a post-deployment processor coupled to the pre-deployment processor and configured to dynamically integrate the single unified code structure into the publisher website as a unified rule set to control at least two predetermined operational aspects of the publisher website based on the first and second set of rules and at least one graphical user interface configured to display to a publisher user of the publisher website at least one operational aspect controlled by the unified rule set.
Rehn (U.S. Pre-Grant Publication No. 2021/0209181) teaches generating dynamic websites that includes a memory storing instructions and at least one processor configured to execute the instructions to perform operations. The operations include receiving an order from a customer device, determining whether the destination address is eligible for delivery by a first time, and based on determining the destination address is eligible, searching a first database to retrieve information of the at least one product. The operations may also include generating a user interface element indicating whether delivery by the first time is possible (the user interface element being configured to modify a website displayed in the customer device) and sending the user interface element to the customer device. Further, the operations may include receiving a response from the customer device and modifying an entry in a second database to indicate the promised delivery date for the product is the first time.
Perez et al. (U.S. Pre-Grant Publication No. 2024/0386197) teaches integrating enhanced model interaction within a website building system. Models leveraged according to embodiments may include trained generative artificial intelligence models that are leveraged to customize structure and content within a website building system. Improved generation of composite prompts leads to improved generation of customized structure and content within the website building system.The reference further teaches “Another quality assurance operation may include reattempting a request (e.g., resubmitting generated prompts and some or all of the data included above as inputs to the AI model) or part of a request. In some embodiments, a request may be modified (e.g., prompts, descriptions, etc.) based on a review of the output map data structure and content, or may include instructions directing the AI model to generate response which is different than the previous response (e.g., in its entirety, or for specific areas in the response).” (Para. [0042]).
Fei et al. (U.S. Pre-Grant Publication No. 2025/0307528) teaches a web page generation method and a device, and relates to the field of front-end development technologies. The method includes: obtaining requirement description information of a web page; inputting the requirement description information into an outline generation model to obtain web page outline information for describing web page logic; rendering the web page based on the web page outline information and a preset web page template; and generating the web page based on the web page outline information and the preset web page template in response to the web page being successfully rendered.
Bista et al. (U.S. Pre-Grant Publication No. 2025/0094455) teaches contextual query rewriting. The techniques include inputting a first user utterance and a conversation history to a first language model. The first language model identifies an ambiguity in the first user utterance and one or more terms in the conversation history to resolve the ambiguity, modifies the first user utterance to include the one or more terms identified to resolve the ambiguity to generate a modified utterance, and outputs the modified utterance. The computing system provides the modified utterance as input to a second language model. The second language model performs a natural language processing task based on the input modified utterance and outputs a result. The computing system outputs a response to the first user utterance based on the result.
Chamberlain (U.S. Pre-Grant Publication NO. 2003/0208369) teaches a user to request information over a network (702) and receive the requested information (708) through one or more information channels. A user, through a client device, may access, through a network, a web page that is hosted on a server. The server, while providing primary information, may further provide an opportunity for the user to request secondary information (710). The user may request access to the secondary information, while maintaining access to the primary information. Included in the request may be a channel selection (704), and associated channel selection information, through which the user wishes to receive the information.
De Ridder (U.S. Pre-Grant Publication No. 2021/0271823) teaches content generation leverages an unsupervised, generative pre-trained language model (e.g., a generative-AI). In this approach, a model derived by applying to given content relevant competitive content and one or more optimization targets is received. Based on optimization criteria encoded as embedding signals in the model, a determination is made regarding whether a template suitable for use as an input to the generative-AI exists in a set of templates. If so, the model embedding signals are merged into the template, or the template itself is transformed using the embedding signals, in either case creating a modified template. If, however, no template suitable as the input exists, the model and other information are input to a natural language processor to generate a generative-AI input. Either the modified template or the generative-AI input, as the case may be, is then applied through the generative-AI to generate an output competitively-optimized with respect to the optimization targets.
Varanasi et al. (U.S. Pre-Grant Publication No. 2025/0086246) teaches optimizing a landing page of a website are presented. The system can receive, from a user device, a first web address associated with a first webpage of the website. The system can process, using the machine-learned assessment model, the first webpage to generate a first landing page score. The system can determine, based on the first landing page score and using a machine-learned optimization model, an actionable suggestion associated with the landing page. The system can cause, on a display of the user device, a presentation of the actionable suggestion.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ROBERT F MAY whose telephone number is (571)272-3195. The examiner can normally be reached Monday-Friday 9:30am to 6pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Boris Gorney can be reached on 571-270-5626. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/ROBERT F MAY/Examiner, Art Unit 2154 8/7/2026
/SYED H HASAN/Primary Examiner, Art Unit 2154
Prosecution Insights