An LLM-based Delphi Study to Predict GenAI Evolution

TL;DR

This work proposes an LLM-based Delphi framework to forecast the evolution of Generative AI in contexts where data are scarce. By implementing a multi-round, prompt-driven Delphi with organizing and responding agents, the method generates structured foresight while mitigating traditional Delphi limitations such as respondent fatigue. The study demonstrates viability and yields insights into geopolitical dynamics, collaboration, economics, ethics, regulation, and societal impact, while acknowledging knowledge cutoffs and biases as key limitations. Overall, the approach offers a novel, scalable reasoning tool for qualitative foresight and policy guidance, with clear avenues for enhancing heterogeneity, external data integration, and real-time knowledge access.

Abstract

Predicting the future trajectory of complex and rapidly evolving systems remains a significant challenge, particularly in domains where data is scarce or unreliable. This study introduces a novel approach to qualitative forecasting by leveraging Large Language Models to conduct Delphi studies. The methodology was applied to explore the future evolution of Generative Artificial Intelligence, revealing insights into key factors such as geopolitical tensions, economic disparities, regulatory frameworks, and ethical considerations. The results highlight how LLM-based Delphi studies can facilitate structured scenario analysis, capturing diverse perspectives while mitigating issues such as respondent fatigue. However, limitations emerge in terms of knowledge cutoffs, inherent biases, and sensitivity to initial conditions. While the approach provides an innovative means for structured foresight, this method could be also considered as a novel form of reasoning. further research is needed to refine its ability to manage heterogeneity, improve reliability, and integrate external data sources.