Browser-Based Chatbot Configuration Using Web AI Metadata
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems require entities to host, train, and configure their own generative AI models for customized chatbot functionality, which is computationally intensive and resource-heavy.
Innovation Solution
Incorporating custom AI metadata within the HTML of webpages enables web browsers to configure generative AI models for chatbot functionality without the need for entities to host or train their own models, using pre-trained language models to provide tailored responses based on webpage-specific metadata.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If entities host, train, and configure their own generative AI models for customized chatbot functionality, then chatbot customization and functionality are improved, but computational resources and system complexity increase significantly
Solution Approach 1:
The patent introduces an intermediary mechanism where custom AI metadata embedded in webpages acts as a mediator between the webpage entity and the generative AI model. The metadata contains instructions and context that enable the AI model to provide customized chatbot functionality without the entity needing to host or train their own model. This intermediary approach resolves the contradiction by achieving chatbot customization through metadata-based configuration rather than full model deployment.
Solution Approach 2:
The patent uses copying by embedding custom AI metadata that replicates the essential configuration and contextual information needed for chatbot customization. Instead of copying the entire AI model (which would be resource-intensive), the system copies only the necessary metadata that enables the AI model to behave in a customized manner, thereby reducing system complexity while maintaining adaptability.
2Adaptability or versatility
If entities host, train, and configure their own generative AI models, then customized chatbot functionality is achieved, but computational resources and energy consumption increase
Solution Approach 1:
The patent extracts only the essential configuration elements needed for chatbot customization from the full AI model training process. By taking out just the custom AI metadata that contains the necessary instructions and contextual information, the system enables chatbot customization without the energy-intensive processes of model training and hosting, thereby resolving the contradiction between adaptability and resource consumption.
Solution Approach 2:
The patent applies preliminary action by pre-configuring the custom AI metadata within the webpage HTML during webpage creation. This preliminary configuration contains all necessary instructions and contextual information that the AI model will need, eliminating the need for resource-intensive real-time model training and configuration when the chatbot is actually used, thus reducing computational resource requirements.
3Loss of energy
If pre-trained language models are used with webpage-specific metadata, then resource efficiency is improved, but the requirement for metadata extraction and processing is added
Solution Approach 1:
The patent merges the custom AI metadata directly into the webpage HTML structure, combining the configuration data with the existing webpage content. This merging approach allows the metadata to be extracted and processed as part of the normal webpage loading and parsing process, minimizing additional complexity while maintaining resource efficiency by using pre-trained language models.
Data Source
AI summary
The technology described herein relates to resource-efficient systems and methods for providing customized chatbot functionality through a web browser, without requiring each entity to separately host, train, and/or configure its own generative AI model. More specifically, custom AI metadata is included in the data package for a web resource. The custom AI metadata is extracted and used to enable customized chatbot features. The web browser identifies the custom AI metadata and uses the metadata to configure a generative AI model that supports chatbot functionality within the web browser. As a result, entities that provide webpages can customize AI-based chatbot functionality by including the custom AI metadata within the HTML content of the webpages.