Multi-AI Response Collaboration via Mediator Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current AI systems at a location operate independently and lack cooperation, leading to suboptimal responses as each device and AI system may have different operating systems, learning techniques, and training parameters, resulting in varying answers to the same user inquiry without determining the best device to respond.
Innovation Solution
A method and system for collaborating among multiple AI devices to generate a superior response by detecting and identifying communicable devices, determining the best responding device based on user location and expertise, and selecting the device to deliver the answer, ensuring optimized response scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple AI systems operate independently at a location, then each system can process requests autonomously, but the response quality varies and there is no mechanism to determine the best device to respond
Solution Approach 1:
The patent introduces a coordination mechanism that acts as an intermediary between multiple AI systems. This mediator collects requests from any device, evaluates which device is best suited to handle each request based on location and expertise criteria, and routes the request accordingly. This resolves the contradiction by providing reliable response quality through systematic selection while managing device complexity through centralized coordination logic.
Solution Approach 2:
The patent creates a universal request handling framework that can process requests from any AI device in the network. The system evaluates multiple devices based on common criteria (user location, device location, expertise) and can route any type of request to the most appropriate device. This multi-functional approach ensures consistent response quality across different AI systems while using a single coordination mechanism.
2Ease of operation
If an AI system responds based on physical proximity only, then the response device is easily determined, but the response may not be from the device with the best expertise or capability
Solution Approach 1:
The patent changes the selection parameters from solely physical proximity to a multi-parameter evaluation system. The system now considers user location, device location, and device expertise/capability parameters. By weighting these parameters (with location having lower importance and expertise having higher importance), the system maintains ease of operation through automated evaluation while significantly improving answer quality by selecting devices based on their capabilities rather than just proximity.
3Adaptability or versatility
If each AI device learns independently based on usage patterns, then each device adapts to local conditions, but devices learn differently and lack cooperation
Solution Approach 1:
The patent merges the independent learning capabilities of multiple AI devices while maintaining their local adaptability. Each device continues to learn from its local usage patterns, but the coordination mechanism enables them to share knowledge about which devices are best suited for which types of requests. This combining approach preserves local adaptation through independent learning while reducing knowledge loss through systematic sharing of device capabilities and performance data.
Data Source
AI summary
A method, system and computer program product for collaborating among multiple electronically communicating AI (Artificial Intelligence) devices for responding to a request includes detecting and identifying devices that are each communicable with a user and electronically communicating and interacting with their respective AI systems. An answer is generated, in response to receiving a request or an instruction by a user at a device of the identified devices, where the answer resulting from collaboration of the identified devices and their respective AI systems. A responding device is determined from the identified devices based on criteria, and a threshold for selecting the responding device. The threshold includes rating, in an order of importance, responding device scenarios for delivering the answer. One of the identified devices is selected to deliver the answer based on the criteria which includes meeting the threshold.