Conversational Surrogate Agents for Large Group Deliberation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in enabling effective real-time conversational deliberations among large groups of networked users, as the effectiveness of conversations decreases with increasing group size, making it difficult to harness collective intelligence.
Innovation Solution
The system divides a large population into smaller subgroups for real-time conversational deliberation, utilizing conversational surrogate agents to observe and distill insights, pass them between subgroups, and express them in natural first-person dialog, while conversational contributor agents provide AI-generated insights and factual information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the group size is increased to harness collective intelligence, then the quantity of participants increases, but the effectiveness of real-time conversational deliberation deteriorates
Solution Approach 1:
The patent divides a large population of users into multiple smaller subgroups, each capable of effective real-time conversation. These subgroups are further segmented into human participants and AI-powered conversational agents. This segmentation allows the system to maintain conversational effectiveness in each small group while collectively harnessing the intelligence of the entire large population through the coordinated interactions of multiple subgroups.
2Adaptability or versatility
If AI-powered conversational agents are introduced to facilitate large-scale deliberations, then the ability to enable coherent conversation among large groups improves, but the complexity of the system increases
Solution Approach 1:
The patent introduces AI-powered conversational agents as intermediaries that facilitate communication between human participants and across different subgroups. These agents observe conversations, distill insights, and transmit them between groups, enabling coherent large-scale deliberation without requiring direct human-to-human communication across the entire population, thus managing system complexity through automated mediation.
Solution Approach 2:
The patent employs multiple instances of conversational agents that replicate the same functional capabilities across different subgroups. Each agent observes, distills, and transmits insights following the same process, creating a scalable template that can be copied across numerous subgroups without proportionally increasing system complexity.
3Speed
If real-time synchronous conversation is used among all participants, then the speed of interaction is high, but the loss of information increases due to the inability to process all inputs coherently
Solution Approach 1:
The patent extracts the functions of observing, distilling, and transmitting insights from human participants and assigns them to AI-powered conversational agents. This extraction allows human participants to focus on generating ideas in their small subgroups while the AI agents handle the complex task of synthesizing and distributing information across the entire system, preventing information loss without requiring all humans to process all information directly.
Data Source
AI summary
The present disclosure describes systems and methods for asynchronous conversational deliberation among a population of human participants. In some aspects, a collaboration server is connected to a memory containing a representation of conversational content that was previously captured from prior human participants, the representation associated with each prior participant including at least one stored idea, insight, view, or assertion expressed by that prior participant and at least one associated numerical measure of sentiment, conviction, confidence, or support. According to some embodiments, an asynchronous conversational process is enabled in which each of a plurality of new human participants engages in real-time dialog with a Conversational Surrogate Agent displayed by their associated computing device, each Conversational Surrogate Agent configured to express to at least one new participant, at least one stored idea, insight, view, or assertion associated with at least one prior participant.