This paper explores the potential of large language models (LLMs) in advancing time series analysis, a critical field for understanding complex real-world systems. While current time series models rely heavily on domain knowledge and extensive tuning, LLMs offer a new approach that can revolutionize the field by enhancing data and model performance, improving prediction capabilities, and enabling more general and interactive analyses. The paper argues that LLMs can serve as central hubs for understanding and advancing time series analysis, promoting efficient decision-making and advancing towards a more universal form of time series analytical intelligence. It highlights the need for researchers and practitioners to recognize the potential of LLMs and emphasizes the importance of trust in these efforts. The paper also details the seamless integration of time series analysis with existing LLM technologies and outlines promising avenues for future research, including the development of LLM-centric time series models and the exploration of multi-agent systems for improved performance and reliability.This paper explores the potential of large language models (LLMs) in advancing time series analysis, a critical field for understanding complex real-world systems. While current time series models rely heavily on domain knowledge and extensive tuning, LLMs offer a new approach that can revolutionize the field by enhancing data and model performance, improving prediction capabilities, and enabling more general and interactive analyses. The paper argues that LLMs can serve as central hubs for understanding and advancing time series analysis, promoting efficient decision-making and advancing towards a more universal form of time series analytical intelligence. It highlights the need for researchers and practitioners to recognize the potential of LLMs and emphasizes the importance of trust in these efforts. The paper also details the seamless integration of time series analysis with existing LLM technologies and outlines promising avenues for future research, including the development of LLM-centric time series models and the exploration of multi-agent systems for improved performance and reliability.