# US 12,010,076 B1 · Increasing security and reducing technical confusion through conversational browser > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,010,076 B1 - Document ID: US12010076B1 - Status: Granted - Kind: B1 - Application number: 18/333,265 - Filed: 2023-06-12 - Granted: 2024-06-11 - Earliest claimed: 2023-06-12 - Assignees: Microsoft Technology Licensing, LLC - Applicants: Microsoft Technology Licensing, LLC - Inventors: Felix Andrew; Ryan Gregory Cropp; Laurentiu T. Nedelcu - Web page: https://patentgrep.com/patent/US12010076B1 ## CPC - H04L 51/02: using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages - G06F 16/3329: Natural language query formulation - G06F 40/35: Discourse or dialogue representation - H04L 51/046: Interoperability with other network applications or services - H04L 51/18: Commands or executable codes - H04L 63/20: {for managing network security; network security policies in general (filtering policies H04L63/0227)} - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Systems and method for providing an application chatbot that provides a conversational interface that receives natural language input from an application user, interprets the user's intent, and uses application-related context for generating and providing a contextually accurate response in a conversation with the user. In some examples, the application chatbot determines an action to perform corresponding to the response and provides an option to perform the action in the conversational user interface. A selection of the option causes the action to be performed. ## Claims ### Claim 1 A computing system for exposing settings of a web browser application, the computing system comprising: at least one processor; and memory storing instructions that, when executed by the at least one processor, cause the computing system to perform operations comprising: displaying a webpage within a web browser application; displaying, concurrently with the webpage, a chatbot interface in the web browser application; receiving a natural language input into the chatbot interface; extracting context from the displayed webpage; generating a prompt for a machine learning (ML) model, wherein the prompt includes the natural language input and the context extracted for the webpage; providing the prompt as input into the ML model; in response to the prompt, receiving an output from the ML model, wherein the output includes executable code for performing one or more actions based on the natural language input; executing a portion of the code to cause the web browser application to perform at least one of the one or more actions; displaying a contextual response, within the chatbot interface, based on the output received from the ML model, wherein the contextual response includes a proposed setting change to a setting of the web browser application; receiving an affirmative response for the proposed settings change; and based on receiving the affirmative response, executing a portion of the executable code to cause the web browser application to implement the proposed setting change to the setting of the web browser application. ### Claim 2 (depends on 1) The computing system of claim 1, wherein the one or more actions include adjusting the setting of the web browser application. ### Claim 3 (depends on 1) The computing system of claim 1, wherein the one or more actions include querying an external data source for additional information. ### Claim 4 (depends on 3) The computing system of claim 3, wherein the contextual response is further based on the additional information from the external data source. ### Claim 5 (depends on 3) The computing system of claim 3, wherein the natural language input relates to security of the webpage. ### Claim 6 (depends on 1) The computing system of claim 1, wherein the one or more actions include at least one of: enabling the setting; disabling the setting; navigating to a new webpage; or performing a document object model (DOM) manipulation of the displayed webpage. ### Claim 7 (depends on 1) The computing system of claim 1, wherein the ML model is a generative AI model that processes the prompt by employing an encoder-decoder structure and self-attenuation mechanisms for multiple layers of a transformer-based neural network. ### Claim 8 (depends on 1) The computing system of claim 1, wherein the setting is a security setting. ### Claim 9 (depends on 1) The computing system of claim 1, wherein the affirmative response is a natural language response. ### Claim 10 A computer-implemented method of enhancing security in a web browser application, the method comprising: displaying a webpage within the web browser application; displaying, concurrently with the webpage, a chatbot interface in the web browser application; receiving a natural language input into the chatbot interface, the natural language input relating to at least one of safety of the webpage or a web browsing session; extracting context for at least one of the displayed webpage or the web browsing session; generating a prompt for a machine learning (ML) model, wherein the prompt includes the natural language input and the extracted context; providing the prompt as input into the ML model; in response to the prompt, receiving an output from the ML model, wherein the output includes executable code for performing one or more actions for changing a security setting of the web browser application; displaying a contextual response, within the chatbot interface, based on the output received from the ML model, the contextual response including a proposed change to the security setting; receiving, within the chatbot interface, an affirmative response; and based on the affirmative response, executing a portion of the code to cause the web browser application to change the security setting. ### Claim 11 (depends on 10) The computer-implemented method of claim 10, wherein the output also includes executable code for querying a data security information source for security data about the webpage. ### Claim 12 (depends on 10) The computer-implemented method of claim 10, wherein the context includes at least one of settings information for settings of the web browser application; an address of a webpage displayed in an active tab of the web browser application; webpage content; one or more entities extracted from the webpage; favorites; browsing history; or webpages of other open tabs. ### Claim 13 (depends on 10) The computer-implemented method of claim 10, wherein the security setting is a tracking setting. ### Claim 14 (depends on 10) The computer-implemented method of claim 10, wherein the ML model is a generative AI model. ### Claim 15 A computer-implemented method of enhancing security in a web browser application, the method comprising: displaying a webpage within a web browser application; displaying, concurrently with the webpage, a chatbot interface in the web browser application; receiving a natural language input into the chatbot interface; extracting context from the displayed webpage; generating a prompt for a machine learning (ML) model, wherein the prompt includes the natural language input and the context extracted for the webpage; providing the prompt as input into the ML model; in response to the prompt, receiving an output from the ML model, wherein the output includes executable code for performing one or more actions based on the natural language input; executing a portion of the code to cause the web browser application to perform at least one of the one or more actions; displaying a contextual response, within the chatbot interface, based on the output received from the ML model, wherein the contextual response includes a proposed setting change to a setting of the web browser application; receiving an affirmative response for the proposed settings change; and based on receiving the affirmative response, executing a portion of the executable code to cause the web browser application to implement the proposed setting change to the setting of the web browser application. ### Claim 16 (depends on 15) The computer-implemented method of claim 15, wherein the one or more actions include querying an external data source for additional information. ### Claim 17 (depends on 16) The computer-implemented method of claim 16, wherein the contextual response is further based on the additional information from the external data source. ### Claim 18 (depends on 15) The computer-implemented method of claim 15, wherein the natural language input relates to security of the webpage. ### Claim 19 (depends on 15) The computer-implemented method of claim 15, wherein the ML model is a generative AI model that processes the prompt by employing an encoder-decoder structure and self-attenuation mechanisms for multiple layers of a transformer-based neural network. ### Claim 20 (depends on 15) The computer-implemented method of claim 15, wherein the one or more actions include at least one of: enabling the setting; disabling the setting; navigating to a new webpage; or performing a document object model (DOM) manipulation of the displayed webpage.