An Innovative Hybrid Soft Consensus Framework Leveraging Generative Large Language Models and Human Expertise

Abstract

Traditional group decision making models have often used artificial intelligence tools, such as fuzzy logic, to act as moderators among multiple human experts tasked with selecting the best option from a set of alternatives. These models facilitated consensus by interpreting and integrating diverse expert opinions. However, the advent of generative Artificial Intelligence models introduces a new challenge: how to integrate these advanced models as active participants in the decision group, alongside human experts. This paper proposes a new hybrid consensus framework that integrates generative large language models as key contributors to the negotiation process. By taking advantage of the unique strengths of both human expertise and AI-driven insights, our framework aims to enhance the robustness and efficiency of group decision making. We explore methods for effectively integrating generative models, addressing potential biases, and ensuring coherent collaboration between human and AI participants. This approach not only enriches the decision making process, but also sets a precedent for future collaborative systems combining human knowledge and artificial intelligence.