From Database to AI Agent: A New Form of AI Participation in Historical Research
10-29-2026

Social Sciences in China (Chinese Edition)

No. 8, 2026

 

From Database to AI Agent: A New Form of AI Participation in Historical Research

(Abstract)

 

Zhang Ping

 

The introduction of large language models (LLMs) and generative artificial intelligence (GAI) into historical research has brought significant changes to the field. In digital historical research, these technologies have addressed the challenges involved in manually constructing databases, simplified software operating procedures, and, to a certain extent, alleviated the difficulties facing the development of digital history. Moreover, they have fostered the design concept of historical AI agents. By replacing databases with knowledge bases, such AI agents enable full-text analysis; by leveraging the autonomous generative capabilities of LLMs and GAI, they simulate historians’ research workflows and ways of thinking to the greatest possible extent, thereby becoming a new type of knowledge system produced in the “digital intelligence era.” To a certain extent, the design of historical AI agents has realized “AI+” and “human–machine collaboration,” opening up new cognitive pathways for historical research. However, this research model cannot replace traditional historical research or digital historical research; it is the diversity of historical research paradigms that provides an important means of promoting the flourishing development of historical research.