Can GPT Help Improve Robo-advisory? The Construction of Robo-advisor for Users with Low Investment Experience Based on LLM

Abstract

Robo-advisors have emerged as a significant innovation in investment management, offering automated financial advice to investors. However, user acceptance remains a challenge, particularly among those with limited investment experience. This paper explores the potential of large language models (LLMs) to enhance the interaction attributes of robo-advisor products and increase acceptance among novice investors. The study contributes to the existing literature by exploring the application of LLMs in robo-advisors, supplementing the exploration of interaction design, and systematically reviewing the service processes of current robo-advisor products. Findings suggest that existing robo-advisor products have room for improvement in interaction attributes and algorithmic mechanisms. Through theoretical exploration, this paper proposes methods for optimizing robo-advisor products by integrating LLMs. In conclusion, this research lays the groundwork for designing robo-advisor products with integrated LLM functionality, offering theoretical references for practitioners and researchers in financial technology. Future research directions include exploring user expectations and conducting controlled experiments to analyze the impact of LLM integration on user decisions.